ReportNo. 18369-ET Ethiopia Review of PublicFinances (In Two Volumes) Volume 11: Appendixesand Statistical Annexes December30, 1998 Resident Mission in Ethiopia Countrv Department 6: Ethiopia, Eritrea, Somalia, Suclan Africa Region Documentof the World Bank Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized
122
Embed
Report No. 18369-ET Ethiopia Review of Public Finances · Africa Region Document of the World Bank Public Disclosure Authorized ... responsibilities for federal ministries are defined
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Report No. 18369-ET
EthiopiaReview of Public Finances(In Two Volumes) Volume 11: Appendixes and Statistical Annexes
December 30, 1998
Resident Mission in EthiopiaCountrv Department 6: Ethiopia, Eritrea, Somalia, SuclanAfrica Region
Document of the World Bank
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
Pub
lic D
iscl
osur
e A
utho
rized
GOVERNMENT FISCAL YEAR
July 7 - July 6
CURRENCY EQUIVALENTSCurrency Unit: Ethiopian Birr
Official Rate: US$ I = Br 7.07 (end of period 1997-98)
ABBREVIATIONS AND ACRONYMS
ASYCUDA Computerized Customs Reporting and RecordingSystem (UNCTAD)
CSA Central Statistical AuthorityESDP Education Sector Development ProgramETB Ethiopian BirrFIRA Federal Inland Revenue AuthorityGDP gross domestic productHSDP Health Sector Development ProgramIPFs Indicative Planning FiguresMEDAC Ministry of Economic Development and CooperationMTEF Medium Term Expenditure FramworkNBE National Bank of EthiopiaNGO non-governmental organizationsPER Public Expenditure ReviewPFP Policy Framework PaperPIP Public Investment ProgramSDPs sector development programsSIPs sector investment programsSNNPS Southern nations, Nationalities and People's StateSOEs state-owned enterprisesSPA Special Program of AssistanceSTB State Tax BureausTIN Taxpayer IdentificationWCO World Customs Organization
1. The purpose of this appendix is to provide a description of the intergovernmentalfiscal relations that are evolving under the Constitutional arrangement that call forRegionalization in Ethiopia. The management of expenditure allocations between theFederal and Regional States is dominated by the tax assignments and expenditureassignments given to each level of government. The question of tax assignments isaddressed in Chapter 2 of this report. In this section, the consequences of the expenditureassignments on the manner in which the Federal Government manages the transfer ofresources from Federal to the Regional states are examined. In so doing, issues forfurther analysis are highlighted. Tackling the implications of each eventuality from thedecentralized system of government in Ethiopia is, however, beyond the scope of thisreport.
The Constitutional Mandate
2. Within the Ethiopian Constitution the powers that are not given separately to theFederal Government are reserved to the regional states.' The Federal Government'sresponsibilities are defined quite narrowly in terms of formulating overall economic andsocial policies, and the establishment of national standards and basic criteria in a numberof sectors. It also has the usual responsibilities given to a federal government in terms offoreign policy (including nationality and immigration rules) defense policy (includinglaws relating to the bearing of arms), monetary policy, inter-State and foreign commerce,and the oversight of inter-regional transport and communications. Specific powers andresponsibilities for federal ministries are defined in more detail by Proclamation No.4111995 on the "Definition of the Powers and Responsibilities of the Executive Organs ofthe Federal Democratic Republic of Ethiopia"; while the responsibilities of the regionalbureaus are set out in Proclamation No 41/1993-on the "Powers and Duties of the Centraland Regional Executive Organs of the Transitional Government of Ethiopia". (This latterproclamation, except for the provisions with respect to central executive organs, stillremains in force although it pre-dates the Constitution.)
3. As a result of its limited mandate, a considerable proportion of the expenditureobligations are the responsibility of the Regional States. In particular expenditure
IArticle 52 (1) of The Constitution of the Federal Democratic Republic of Ethiopia, December 1994.
2 Appendix I
Box Al.I: The Budget Subsidy formula for capital allocations in 1994195
TRi = TRx(0.3OxPOPi/POP + 0.25xli/IxPOPi/POP+0.2OxSRBi/SRB + 0.15 Ki/K+0. 10 Ai/A)
Where:TRi Transfer to region iTR Total of transfers over all regionsPOPi - Population of region iPOP - Total population over all regionsIi = The so-called "I-distance" indicator of regional underdevelopment
(expressed as region i's fraction of the "sum" of underdevelopmentover all regions)The average "I-distance" underdevelopment of the regions (expressedas a sum over all regions)
SRBi Share that region i expects to cover with its own tax revenue of its totalbudget ceiling
SRB - Average share over all regions of budget ceiling covered with own taxrevenue
Ki 1992/93 capital expenditure of region iK = 1992/93 total capital expenditure over all regionsAi Area of region iA Total area over all regions
obligations in the area of rural roads, education (except higher education institutions), andhealth (except referral hospitals and research centers) fall upon regional governments.However, the tax-raising powers of the Federal Government are much greater than theRegional States, with the result that the Regional States rely to a considerable extent ontransfers from the Federal State. The evolving nature of the relations between the FederalGovernment and the regional states are summarized in Box Al.1. In the discussion
below we focus on the relationship established for fiscal year 1997/98.
The Determination of Regional State Budget Ceilings and Budget Subsidies in1997/98
4. The transfer of resources from the Federal State to the Regional States is a
judicious mix of a transfer formula combined with some subsequent adjustments to take
account of particular circumstances that cannot be captured in the formula. The amount
of funds that is transferred to each region in the form of a "Budget Subsidy" 2 is
determined on the basis of a number of steps as follows:
2The term "Budget Subsidy" is used in the Financial Administration Proclamation No. 57/1996 to describe"the annual budgetary amount provided by the Federal Government to a Regional Government". Inthe Fiscal Federalism literature this type of transfer is more commonly referred to as a "grant". In thetext below we maintain the Ethiopian practice of using the phrase subsidy.
Federal-State Fiscal Arrangements in Ethiopia 3
Table A1.1: Key Statistics On Ethiopia's Regional States
National Regional Region Population Area in Number Number NumberStates/ Capital 1996/97 '000 km2 of of of Special
Source. CSA, 1996/97: compiled from reports submitted by Regional Planning and Economic DevelopmentBureaux, 1996/97
e First, the overall resources available for general government (federal andregional) public expenditure is determined on the basis of historical figuresand forecasts in GDP and government receipts (made up of governrnentrevenues, and external assistance3 and loans).
Second, the overall resources available for general government expenditureare then divided between federal and regional expenditure on the basis of theexpenditure assignments at each level of government.
Third, the overall envelope of resources for regional governments is thendivided between the regional states on the basis of a "budget subsidyformula" which determines the total "budget ceiling" for each regionalgovernment.
* Fourth, the budget ceiling for each individual regional state is then reducedby the deduction of forecast "own revenue" to arrive at a budget subsidy. Itis this budget subsidy which is set out in the Federal Government's BudgetProclamation.
* Fifth, the budget subsidy is then subdivided into that amount that ischanneled through the federal Central Treasury, and that amount that ischanneled through external assistance and loans. The result is that thetransfers from the Federal Government to the regional states is made up onlyof the amount from the federal Treasury. The deduction of foreign assistance
3 The terrn foreign assistance is used to refer to foreign grants as opposed to foreign loans.
4 Appendix I
and loans from the budget subsidy is commonly referred to as the "budgetoffset".
5. The process from budget ceiling through to budget subsidy, to central treasurysubsidy can be seen for the 1997/98 year in Table 1.9, Chapter 1, Volume I. There arefurther steps that are carried out by the regional states, which are not considered here.For example, in the SNNPS it is understood that a formula is applied to allocate fundsfrom the regional government to zonal and special-woredas.
6. Of particular interest is the fact that there are sources of revenue or receipts thatRegional States can make use of for expenditures which do not appear to be included inthe methods outlined above. Of particular importance are (i) the regions' rights andresponsibilities to raise their own non-tax and tax revenue which they appear to be able todisburse with the minimum of influence from the Federal Government; (ii) the way inwhich the regions' share of the joint federal-regional taxes are calculated and disbursed,and the fact that these joint revenues are not included either in the reduction of the budgetceiling, nor in the formula; (iii) the manner in which regions can carry forward cashreserves from one year to the next in a manner that it outside the system of budgetarycontrols; and (iv) the role of borrowing by regions, which does not yet appear to haveformally established procedures.
The Subsidy Formula
7. There has been a succession of increasingly refined formulae in Ethiopia todetermine first, "theoretical" allocation of the federal transfers to the regions,4 and thensecond, the "actual" allocation on the basis of the theoretical allocation along withadjustments considered necessary to reflect factors that the formula does not fully cover.
8. Fiscal year 1994/95. Different principles were applied for allocating an currentexpenditure transfer to the regions and for allocating a capital expenditure transfer. Forcurrent expenditure no precise formula was used, but five broad criteria were used: thenumber of woredas and zones in each region, the 1993/94 transfer allocations, the lengthof rural roads, the number of state agricultural demonstration centers and the 1993/4expenditure for education and public health.
9. For the capital expenditures, the following formula was used was a weightedmixture of population, I-distance, regional tax effort, capital expenditure in the previousyear, and the regions area (see Box Al.2). The 1-distanice indicator was based on eightdifferent factors - length of rural roads, share of rural population in total population, percapital industrial production, per capita crop food production, density of telephone lines,number of post offices, hospital beds per head of population, and number of pupils inprimary (or elementary?) schools per head of population. The I-distance calculated using
Giorgio Brosio & Sanjeev Gupta, "Ethiopia", in Teresa Ter-Minassian, ed., Fiscal Federalism in Theoryand Practice, Washington, D.C., International Monetary Fund 1997, 504-526.
Federal-State Fiscal Arrangements in Ethiopia S
Box Al.2: The Budget Subsidy formula in 1995196
TRi = TR x (1/3 x POPi/POP + 1/3 x liHA x POPilPOP + 1/3 x SRBi/SRB)
Where the symbols are as in Box A3. 1.
this formula included an element to account for the correlation between the variables soas not to "double-count" the extent of underdevelopment in each region.
10. Fiscal year 1995/96. In 1995/96 the formula was extended to cover bothrecurrent and capital expenditures. Accordingly, the transfer is fully fungible as to theregion's rights to use it for either recurrent or its capital expenditures. Compared to1994/95 the formula was simplified so that it excluded the elements of capitalexpenditure and area. As a result the weights were altered, with one-third given topopulation, one-third given to the I-distance (itself weighted by population) and one-thirdgiven to tax effort. The formula is summarized in Box Al.2. The 1-distance indicatorwas still calculated using a method to account for the correlation between thedevelopment indicators so as not to "double count" the level of underdevelopment.
11. Had the formula introduced for 1995/96 been applied in full, the transfer shares ofthe wealthiest regions of Addis Ababa and Dire Dawa had substantially increased incomparison to their prior shares. To avoid this with a goal of limiting changes in transferamounts, a discretionary correction to the transfers was made to decrease the changes in
5regions' transfer shares between the two fiscal years.
12. Fiscal Year 1996/97. A number of changes were made to the transfer formula in1996/97. The I-distance indicator was simplified in a number of respects. First, theweighting of the I-distance by population was removed. Second, instead of being basedon eight factors the I-distance indicator was based on five factors-number of clinics,number of primary schools, number of telephone lines, electricity consumption, andlength of roads. Third, the I-distance was calculated as a simple linear combination of theabove five factors, instead of including a method to take account of the correlationbetween these factors. The resulting formula is set out in Box Al.3.
Box Al.3: The budget subsidy formula in 1996/97
TRi = TR x (1/3 x POPi/POP + 1/3 x li/l + 1/3 x SRBi/SRB)
Where the symbols are as in box x.x above Box Al.I
Brosio & Gupta, art. cit.
6 Appendix 1
13. The additional change made in 1996/97 was that the "budget offset" wasintroduced for the first time in this year. It has been continued in 1997/98.
14. The budget subsidy formula that was used in 1997/98 is set out in Box 1.4.Although the formula set out above appears complex, it represents an improvement onprevious years. Noticeably, the weight given to population has increased from 33.3percent to 60 percent, and the Development Index has been simplified so that it includessix elements that appear to be closely related to the strategic objectives of thegovernment, rather than the eight rather heterogeneous variables used in 1995/96.
15. The purpose of the three grant elements can be characterized as follows. The firstelement in the transfer formnula aims fundamentally at correcting the considerable federal-regional fiscal imbalance, which involves a federal share of about 60 percent of the totaltax revenue despite the fact that about 60 percent of all expenditures take place at aRegional level in line with the fundamental devolution of responsibilities to the regions.Since this element is also based on population this element involves aspects of universalprovision of services for all citizens.
16. The second element aims at promoting the allocation of public expenditurestowards critical development needs. Its constituents are generally in line with theemphases in key national development programs-education, health, roads, telephones,electricity, and water.
17. The third element of the formula aims at giving the regions incentives to raisetheir own revenue. As own revenue as a percentage of the budget increases, then theregions share of total resources available to Regional States increases. Thus there areincentives to increase collection rates (as well as expand the tax base and increase taxrates in compliance with national legislation).
18. There are some reasons to believe that the third element in the formula is unlikelyto have a strong influence on own revenue mobilization. In the first place, the weightgiven to this element is not particularly high, and in fact has decreased from the weightgiven in previous years (see Annex 1). Second, the fact that a regions own revenue isdeducted from the "budget ceiling" before determining the budget subsidy, suggests thatthe benefits of a successful attempt to increase own revenue in the previous year may beoffset in the current year if it is expected that successful revenue mobilization willcontinue. In fact, the net effect of these two factors would be to reduce the overall budgetsubsidy.
19. However, the technicalities in this area may be of minor importance. In practice itappears that if a region raises more revenue than forecast, this revenue can be used toincrease expenditure in the current year, or by carried forward to the next year where itcan similarly be used to increase expenditure. For example, [need to check case ofOromia and Addis Ababa-they appear to have been able to utilize cash balance fromprevious years without it affecting their budget ceiling or their budget subsidy]Furthermore, Regions appear anxious to reduce their heavy dependency on the Federal
Federal-State Fiscal Arrangements in Ethiopia 7
Box AI.4: The Budget Subsidy formula in 1997/98
TRi = TR x (0.60 x POPi/POP + 0.25x Di/D + 0.15 x [(REVi, t-l/BUDi, t-l)/(REV t-l/BUD t-I)1)Where:TRi = Transfer to region i [need to check the term used here: is it budget ceiling, gross
grants or whatever].TR Total of transfers over all regions.POPi Population of region i.POP Total population over all regions.Di = A composite inverted development index, which for region i is equal to
6/(Ei+Hi+Ri+ELi+TEi+Wi). The official title for this index is "Index for thecombined indicators of the level of development of a region".
D Sum of the values of the Di over all regions.Ei Index for combined indicators for education in region i (made up of three
equally weighted sub-components: average class size, pupil-teacher ratio,primary and secondary school participation rate as percentage of age cohort,each weighted equally).
Hi Index for combined indicators for health in region i (made up of six equallyweighted sub-components: number of clinics, number of doctors, number ofnurses, number of health assistants, number of hospital beds, and mortality rateof under-fives).
Ri Index of road density in region i (road length per km2).ELi Index of electric power sales region i (KWH per thousand people).TEi = Index of telephone line density in region i (telephone lines per thousand people).Wi Index of coverage of safe drinking water in region i (percentage of people with
safe drinking water).REVi, t- I Planned own revenue raised by region i in the year prior to the year for which
the transfer is being calculated (since it is "own" revenue it excludes theregion's share of joint revenue with the Federal Govemment).
BUDi,t- I = Planned budget for region i in the year prior to the year for which the transfer isbeing calculated.
REVt- I = Sum of planned own revenue over all regions prior to the year for which thetransfer is being calculated.
BUDt- I = Sum of planned budget over all regions over all regions prior to the year forwhich the transfer is being calculated.
Government. in order to ensure greater independence in the management of their financialresources. It would seem, therefore, that for reasons that lie outside the transfer formulathere are sufficient incentives for own revenue generation.
20. There have been many changes in the transfer fornula over recent years (see BoxAl .4), and we might therefore expect there to be further changes in the future. Steps thatmight be taken include increasing the own revenue element in the formula to include theregion's share of joint Federal-Regional revenues; the inclusion of a measure of capacitybased upon the number of civil servants per relevant office; altering the road densityformula to reflect the different types of road in each region; and perhaps a change in theweighting system within the Development Index to better reflect national developmentprogram objectives.
8 Appendix I
Table A1.2: Regional states share of total budget allocation, 1993/94 to 1997/98
21. One interesting feature about all the discussion of the formula and its application,is that in practice its influence on budget allocations does not appear to be particularlystrong. Despite the many changes in the formula, there has been no dramatic change inthe share of resources given to individual regions (see Table A1.2). The trends, if any,that can be detected are small declines in the budget allocation to the larger regions ofTigray. Amhara, Oromia. and SNNPS; while at the same time there is a small increase inthe allocations to the less advanced Regions or Afar, Somalia, Beneshangul/Gumuz, andGambella.
22. One reason why the shares given to each region have not changed dramatically isthat Ethiopia has followed common international practices in that changes following fromrevisions of the grant formula have been accompanied with particular ad hoc measures tosoften the immediate impact of the changes upon the grant recipients. Importantly, inEthiopia the transfers to be received by the fiscally strong administrative councils ofAddis Ababa and Dire Dawa have been adjusted downwards to provide greater resourcesfor the fiscally weaker regions. Some degree of ad hoc adjustment is generallyunavoidable in grant systems and frequently also desirable for flexibility. However,transparency in overall public expenditure management would suggest that ad hocadjustments should be reduced over time and some options in this area are suggestedbelow.
The Budget Subsidy Formula and The Promotion of National Priorities
23. The discussion on fungibility above indicates that in effect the system of CentralTreasury budget subsidies to each region operates as a "block grant". In contrast, thedonor resources, whether in the form of assistance or as loans, have their use determinedbefore the budget is concluded. They, after all, are the result of discussions held betweenthe donor and the region. The resources cannot be transferred from one use to another
Federal-State Fiscal Arrangements in Ethiopia 9
Box Al.5.: The Ethiopian Grant System from the Perspective of Grant Theory6
Inter-governmental transfers can be categorized according to a number of criteria. We can identify threemain categories:
Unconditional grants: unconditional grants (also known as block grants) are made available on the basisof a number of criteria, but the recipient government is free to utilize the funds according to its ownpriorities and objectives. The unconditional grant is often made available to the other tier of govemmenttaking into account expenditure needs, the tax base, and the tax effort. This form of grant does not requireany ex-post verification beyond the customary assessment of an external auditor.
Conditional grants: conditional grants (also known as specific purpose grants, or categorical grants) aremade available for a particular purpose. These grants are often lump-sum in the sense that they are madeavailable in one block. Ex-post verification of the use of the grant is required to ensure that it has beenutilized for the purpose intended.
Matching grants matching grants (often seen as a sub-category of conditional or special purpose grants)are made available if a lower level of govemment devotes some of its own resources to the purposeintended in the matching grant. The grant may be performance-based, emphasizing the recipient'sultimate outputs, or it may be cost-based, emphasizing actual expenditures. For example, in a cost-basedsystem, if a regional govemment spends 100 units on, say, education, a matching grant would makeavailable, say 10 or 20 percent of this amount from the central government. Matching grants may be open-ended in the sense that there is no limit on the amount that can be made available from the centralgovernment, or closed-ended in the sense that there is a ceiling in terms of the amount of funds available.Matching grants require a system of ex-post verification of expenditures, and since the payments are alsomade ex-post, they also often involve financing costs since the regional govemment can incur expensesbefore it is reimbursed.
Other grants: other forms of grant exist. For example, there are sometimes grants that arise fromcompetitions for a specific pool of money, or one-off capital grants for specific purposes.
From this perspective the Ethiopian system is primarily one of unconditional grants based upon commonnon-discretionary criteria of expenditure needs and tax base and tax effort. In Ethiopia theunconditional grant is strong in the per capita part of the fonnula. The third element in the formula is alsoone of a general grant, aiming at an incentive to the regions towards revenue financing of their budget. Afeature absent in Ethiopia, but common in many other countries, is horizontal fiscal balancing, or partialequalization of the per capita tax base of the different regions. The emphasis in Ethiopia is on encouragingthe raising of taxes. In addition a grant aspect absent in Ethiopia is the conditional (specific purpose orcategorical) or matching grant.
without the prior agreement of the donor, if only because the disbursement mechanismsare outside the normal federal-regional channels. In some cases donor funds may only bedisbursed if the region makes available some of its own matching funds for the projectpurpose. In this sense the resources received from donors can be seen as "conditionalgrants" (where the region does not have to make available some of its "own" resources),or "matching grants" (where the region does have to make available some of its "owneresources and where the matching coefficient is, unusually, greater than one). (For anoutline of different approaches to inter-governmental transfers see Box A1.5.)
10 Appendix ]
24. Given the objectives of government, a number of options can be discussed forimproving the grant system. We can consider the overall nature of the grant system andthe specific features of the current grant formula.
25. If we look first at the overall nature of the grant system, it is apparent that weshould be guided by the need to strike a careful balance between national objectives todevolve power to the regional states (in line with the Constitution), and at the same timepursue particular objectives in particular sectors (such as set out in the sectordevelopment programs). Of importance in this respect is the Ethiopian Government'swish to encourage donors to "pool" resources with the central government's resources(presumably in the Consolidated Fund), and then distribute these resources to the regionalstates in line with the current transfer formula. An objective of this nature immediatelyraises issues about the unconditional grant that is made to regional states, and the fact thatdonors might wish to see their resources focused on particular sectors, with some form ofex-ante understanding that the budgetary system is designed to deliver this result.
26. Options in this area include:
* keeping donor funds outside the transfer system-in much the same manneras at present, where donor funds operate de facto as conditional loans andgrants; or
* incorporating donor funds fully into the Federal Government's transferarrangements, but relying on the dialogue surrounding the SectorDevelopment Programs and the budget making process to deliver the resultsthat are required; or
* incorporating donor funds fully into the Federal Government's transferarrangements, but altering in some way the manner of its operation throughthe use of conditional or matching grants, and thereby giving greaterassurance that the results required will be delivered.
27. In examining these options we need to consider the advantages and disadvantagesof different approaches. The advantage of the current unconditional grant system is thatit minimizes the transaction costs associated with inter-regional transfers and at the sametime, through maximizing the fungibility of resources, ensures that regions are able toadapt their budgets effectively in line with local priorities (which, after all is one of thevirtues of decentralized government). On the other hand, as with any unconditional grantscheme, it is weak in the internalization of inter-recipient externalities where they exist(such externalities may be involved, for instance, in building some rural through roads,the maintenance of food security and the control of contagious disease such as the AIDS)and it does not facilitate the explicit pursuit of national goals.
28. This is a difficult and complex area, particularly given the overall objectives ofthe Ethiopian government. It is an area where the options need to be examined carefully,and it is perhaps wise to leave a full assessment of these aspects to future work to becarried out on the regionalization process in Ethiopia.
Federal-State Fiscal Arrangements in Ethiopia 11
29. If we turn our attention to the specific features of the existing grant regime we canmake a number of observations. The weight given to population in the first element ofthe formula has the merit of simplicity. However, it might be worth considering ifinternational best practices could be applied to adjust this element to take account offeatures such as low population density.
30. The third part of the transfer formula based on a regions success in raisingrevenue undoubtedly gives the regional sates incentives to fund their budget with theirown tax revenue. At present a regional states share of revenue from the joint federal-regional taxes is not included in determining the size of the budget subsidy, suggestingthat this is an area that could be examined to try an better equalize the regional state'saccess to resources.
31. Furthermore, it is worth considering whether this third element in the transferformula could be developed to incorporate an element involving equalization of theregions' standardized tax base. A formula element for partial equalization of the tax basewould have the advantage that it would abolish the current awkward practice ofsubtracting the region's own revenue from the gross subsidy first calculated by theformula. The equalization scheme should operate within a scheme where the tax base isheld constant despite idiosyncratic local taxes and the tax rate is also held constantdespite that regional tax rates may vary. The standardization would involve animprovement upon the present practice where standardization of those aspects is notexplicitly present.
32. The second element of the formula covering the level of development is the onearea in the formula where the Federal Government can indicate what it considers to be itspriorities. As currently structured this element with its focus on education, health, roads,power, telecommunications and water, is clearly based upon current priorities. Otheraspects that may become important include "capacity building" at regional level, andaltering the weight given to particular elements within the formula, perhaps emphasizingparticular objectives such as the education of women and girls7 which would promote thehigher social rate of return expected from expenditure in these areas. In addition asattention is turned to developing further the sector development programs in theeducation and health the weighting that is given to these factors within the formula couldbe increased.
APPENDIX 2
ETHIOPIA'S CIVIL SERVICE
INTRODUCTION
1. The success of any of the changes in the institutional arrangements for publicexpenditure management that are currently being carried out, along with those that areplanned and suggested for the future, will depend critically on the quality and quantity ofcivil servants. In fact, some regions have already announced efforts to devolve financialmanagement responsibilities towards lower levels of government and also to basis servicedelivery units. As such these efforts are very welcome.8
2. The Ethiopian civil service is highly committed and dedicated. In order for theongoing SDPs in education, health and roads to be implemented successfully and in atimely fashion at the regional levels as envisaged, the Government is in the process ofdealing with the severe capacity differences that exist among the regional states inEthiopia. This appendix provides a comparison of the size and professional qualificationsacross the Regional states in Ethiopia. It gives an idea of the task ahead whenformulating appropriate capacity building and training programs that are geared towardseventually having in place a civil service across the country that facilitates private sectoractivities through appropriate delivery of public services and effective implementation ofthe envisaged sector development programs over the medium term.
The Size and Composition of Civil Service
3. The reported number of permanent civil servants in Ethiopia is about 310,000, ofwhich about 265,000 - or about 85 percent - work in the regional states (see Table A2.1).The number of civil servants per 1,000 of population is equal to the internationally lowfigure of about 5 in 1997. For the regional states the figure is about 4.4 (See Table A2.2below). Ethiopia certainly does not belong to the many countries with an overstaffedcivil service. On the contrary, the concern in Ethiopia may be that the number of civilservants is unnecessarily low given the legitimate tasks of the country's public sector.That number has however been increasing in the recent years partly due to a generaldiversification of public administration tasks (see Table A2.1). Although availablestatistics are not complete, there has been an increase in civil servants by about 40 percent
See for example, Stephen Peterson, "Generic Issues in Financial Management for the CentralGovernment," p. 245-268 in Social Sector Review/PER III, May 1997 Vol. 2, and Yishak Mengesha,"The Process of Budget Preparation and Utilization at the Social Service Provider Level", p. 305-334in Social Sector Review/PER 111, May 1997, Vol. 2.
14 Appendix 2
between 1991/92. Meanwhile, the expenditure on wages and salaries for Ethiopia's civilservice have not increased as a proportion of total expenditure over the last few years.
4. In 1996, the three most populated regions in Ethiopia were slightly below thenational average in terms of civil servants per 1,000 of population. The regions with thelargest number of civil servants per head of population were the urban areas of Harare,Addis Ababa and Dire Dawa, as well as the small, least developed regions, ofBenishangul-Gumuz and Gambela. The higher number of civil servants does not,however, compensate for the severe capacity limitations that exist in the least-developedregions of the country. The Government is making special efforts to provide appropriatetraining and capacity building to these regional administrations.
Table A2.1: Civil Service Staffing, 1991/91-1996/97(end-year figures) 1994/95 1995/96 1996/97
Total 293,452 287,716 308,950Source: Personnel Statistics, 1988 EFY, Federal Civil Service Commission, May 1997Yotes: Figure taken froan June /995 Information
-Figzre taken fronz jne 1995 information?Siu'ee the information is not complete the toalfigure is reported by buireau
blanks inidicate thlat no officialfgigre is avalaible
5. If one were to look at professional qualifications in terms of the number of civilservants witlh a diploma and above, the regions of Beneshangul/Gumuz and Gambelawere the most under-provided. In both cases the share was about 1O percent, compared toa figure as high as 25 percent in the case of Addis Ababa, and an average for all regions(for which statistics are available) of just under 16 percent.
6. It is likely that the tasks of "administration and general services" take a highershare of the regional civil service staff in the smaller less populated regions, such as Afar,Benishangul-Gumuz, Afar, and Gambela. Naturally, this result would arise from thehigher overhead costs associated with such regions, suggesting that there are goodgrounds for these regions to receive a relatively larger share of the federal grant (a reasonthat would be further strengthened given the low tax base in these regions-see Chapter2).
Ethiopia's Civil Service 15
Box A2.1: Human Resource Management in the Civil Service Reform
There are eight components within the Human Resource Management component of the Civil ServiceRefonn which are summarized below.
Legal Framework: adoption and implementation of comprehensive Proclamation, Regulations andDirectives.Human Resource Management Information System: development and implementation of computerizedmanagement information system covering key aspects of human resource management in the civil service.Job classification and grading: development and implementation of job classification and grading systemthat allows for efficient and flexible deployment of staff whilst ensuring fair recognition of work.Remuneration and exit interviews: development of labor market surveys and arrangements for exitinterviews to provide infornation on the ability of the civil service to recruit, motivate and train necessarystaff.Staff assessment, promotion, and time management: implementation of an assessment and promotionsystem that allows fair, honest, and open evaluation of staff and their development needs, including abilityto perforn at a higher level.Recruitment, selection, transfer: implementation of improved arrangements for the recruitment,selection, and transfer of staffHuman resource planning: implementation of arrangements to ensure institutions plan their humanresource requirement on the basis of securing government objectives.Development of human resource management professional service: development of necessary humanresource manag,ement skills within the civil service.
Source: Etl2iopian Civil Service Reform: Four Year Plan, 1997-2000, May 1997
7. In recognition of the importance of the civil service in delivering services, andensuring the proper use of public funds, two major initiatives have been taken in recentyears. The first is the establishment of the Civil Service College; and the second is theHuman Resource Management component of the ongoing Civil Service Reform program.The World Bank and UNDP, along with other donors, are providing grant assistance inthe implementation of some of the components of this program.
8. Civil Service College. The Civil Service College which is based in Addis Abababegan functioning in January 1995, and was established as an autonomous highereducation institution under Proclamation No. 3/1996 of 15 February 1996. It has a multi-national permanent teaching staff, and has three main faculties-of Law, Technology,and Business & Economics. Most of the courses are of three years duration, with themajor areas being Law, Economics, Accounting, and Municipal Engineering. There arealso shorter three month Certificate Courses in Accounting. There are also plans tointroduce a course in Public Administration shortly.
9. The existing needs for strengthening the professional civil service seem to be welladdressed in the overall orientation of the training provided by the Civil Service College.However, the scale of the task facing the institution should not be underestimated. As ageneral guide, it is often suggested that about 5 percent of staff time should be spent in
16 Appendix 2
Table A2.2: Civil Servants professional status (June 1996) and training (1994/95-1997198)
No. who PercentNo. of prof. Percent of have who
Permanent civil prof. civil Those who received receivedNo. of civil servants servants attended training in training in
permanent servants with with ECSC Accounting Accountingcivil per '000 diploma diploma (1994/95- (1994/95- (1994/95-
servants pop. and above and above 1997/98) 97/98) 1997/98)Tigray 17,338 5.2 2,372 13.7 250 70 28.0Afar 4,067 3.6 na na 49 11 22.4Amhara 59.345 4.0 9,802 16.5 298 88 29.5Oromia 85,660 4.3 11,642 13.6 283 77 27.2Somali 5,360 2.7 na na 200 66 33.0
DawaTotal 245,426 4.4 37,437 15.9 1,719 455 26.5Soutrce: Regional Dcvelopmeni in Ethiopia, MEDAC, February 1998
training.9 If we take say half of this figure, then this is equivalent to about 6,000 person-years of staff training per year. If we half this figure again to take account of trainingoutside the civil service college we would have a figure of 3,000 person-years-still one-third higher than the amount of training that has taken place in the civil service collegeover the last three years. The civil service, therefore, faces a daunting task ahead in termsof training its staff.
10. If one looks at individual regions, Benishangul-Gumuz and Gambela showconsiderable scope for strengthening the professional civil service in the near future. Inthe area of public expenditure management, the total number of students from each regionwith an accounting major in the annual courses of the Ethiopian Civil Service College in1994/95-1997/98 can be seen in Table A2.2. The accounting major is very stronglyconcentrated in aspects of public expenditure management including budgeting. TheEconomics major-not shown in the table-is targeted more towards the Planning andDevelopment functions for the regions. The number of accounting students from theregions can be related to the total number of students receiving training, with the largest
9This is the figure used in the UK civil service. Some private institutions have a much higher figure forstaff training-Anderson's, the management consultants have a rule of using 7 percent of staff time intraining
Ethiopia's Civil Service 17
share of public expenditure management training expected in the Dire Dawa, Somali,Harari, Amahara, and Oromia regions. Overall about one quarter of all those trained havebeen in the area of accounting.
11. The work that is planned under the HRM component is still at the preparatorystage. However, two central issues appear in all this work: the remuneration of staff; andtraining of staff. It is widely acknowledged (although official statistics are not availableat the moment) that the Civil Service is losing its high quality staff in a manner that workin key areas is being and could be hampered in the future. Some Regional States appearto have taken action in recent years to reward particular categories of staff with higherwages. However, there is an urgent need to address carefully and quickly a number ofpriority areas to ensure that government objectives as a whole can be pursued. Ofparticular note will be the remuneration of those working in the key sectors (andparticular professions within those sectors) of the priority SDPs/SIPs, namely roads,education, and health. In addition those areas where there are shortages of staff such asthe Customs Authority (which is running at 55 percent below the staffing levelrecommended) and the Privatization Agency (where there are apparently a total of 15professionals covering privatization and the return of illegally confiscated property).
12. In responding to the needs in this area some form of remuneration package needsto be prepared that brings together government objectives and market conditions. In linewith this approach key features would include: (i) wages that reflect the market positionof the staff involved, so that high quality staff can be retained and recruited; (ii) noacross-the board pay increases to all classes and categories of staff and instead focuses onparticular professions in short supply, for example, lawyers and contract specialistswithin both the Ethiopian Roads Authority and the Privatization Agency; (ii) recognizesthat there will probably be a need for the narrowing of differentials based on years ofservice-new recruits may need to be paid similar wages to those who have worked formany years if that reflects market conditions.
13. The second critical factor is that of capacity building and training. Much workhas been carried out over that last few years in enhancing the capacity of civil servants,most notably through establishment of the Civil Service College. In addition specificprograms of the government, especially in the sector development programs, areaccompanied by programs to improve capacity. However, the task is extremely large.There are around 300,000 civil servants and ensuring that all are aware of, say, thereforms in expenditure management and control will be extremely demanding: andensuring that there are sufficient skills for those who have to carry out the reforms will beeven more demanding. Considering that other reform programs in roads, education, andhealth will also require training and capacity building serves to reinforce this point.Furthermore, taking into account that out of the 300,000 or so civil servants about 1,700have attended the Civil Service College, serves to remind us that even those highlyimportant reforms have only started to address the larger issue of capacity building.
APPENDIX 3
ETHIOPIA'S RECENT TAXATION AND TARIFFREFORMS
1. The Ethiopian tax system has become more favorable to the private sector inrecent years. This is being achieved through a more or less continuous process ofreforms, based in part on findings of several background studies that have been conductedwith the donor cooperation and assistance at the initiation of the Ethiopian authorities.This appendix summarizes the main policy and institutional taxation-related reforms thathave been undertaken by the Government in the 1990s.
2. Tax Reforms. Specifically several maximum tax rates have been reduced: themaximum rate for personal income tax has been reduced from 89 percent prior to 1992 to40 percent in 1994, the maximum for business income and profit has been reduced from89 percent to 40 percent in 1994 for unincorporated firms and 35 percent for corporations,the maximum rate for mining income taxes was reduced to 35 percent, several excise taxrates were reduced, the sales tax for low income service providers was reduced from 10 to5 percent, the sales tax on imports was reduced from 24 to 12 percent, and the maximumimport tariff rate was reduced from 200 percent to 50 percent. (See Table A4. 1.) Thedeadline for the payment of the sales tax has been increased to a more manageable 7working days. The tax regime has become simpler to administer as well: the number oftax rate brackets has been reduced for the personal income tax, the business income taxand for import tariffs. The federal government also plans to consolidate the four taxes oncoffee exports into one uniform tax with a rate of 6.5 percent. There were only a few set-backs from the private sector perspective. The excise tax for Landrovers, Jeeps and other4 wheel drive vehicles was increased from 20 to 40 percent and the capital gains tax wasintroduced.
3. Tariff reforms. The recent reforms documented above will help reduce businesscosts and increase consumer purchasing power. Yet, there is a need for continued reformin the case of imports. The Government is in fact considering a further reduction in themaximum tariff rate to 40 percent by the end of 1998. There is a strong case forcontinued tariff reforms. In particular, the unweighted average of 21.5 percent militatesagainst competitive production and exports. (See Box A4. 1) The high tariff rates add toproduction costs across most sectors of the economy. This makes Ethiopian products lessdesirable in the market place both at home and overseas. High tariffs also tend toappreciate the real exchange rate, making Ethiopian exports even less attractive. Some ofthe revenue losses from reductions in tariff rates can be recovered: the number ofexemptions can be sharply reduced, zero tariffs can be replaced by low tariffs between 1and 3 percent, net tariff collections can be improved by reducing the administrative costthrough fewer bands, and domestic taxes can be increased.
89 maximumAmend. 30/1992 Income up to Birr 150 9 10 minimum
50 maximumAmend. Income up to Birr 120 5 10 minimum107/1994 40 maximumBusiness Income Tax (Incorporated Firms)Pre-1994 None 1 50 all income levelsAmend. None 1 40 all income levels107/1994Amend. 36/1996 None 1 35 all income levelsBusiness Income Tax (Unincorporated Firms)Pre-1994 Income up to Birr 300 17 3 minimum
59 maximumAmend. Income up to Birr 1.200 5 10 minimum107/1994 40 maximumIncome Tax on Mining53/1993 Artisinal mining 1 45 large scale
35 small scaleProc. 23/1996 1 35 for all operations Allows Government
to acquire equity upto 2% (down from10%) free of cost inany large scaleinvestment.
Income Tax on RentalAmend. 62/1993 Birr 1,200 6 10 minimum Reintroduction of
45 maximum rental tax afternullification duringDerg regime.
Capital Gains Tax
Proc. 108/1994 First Birr 10,000 1 30 federal First introduction of10 state capital gains tax.
Sales Tax on Goods and ServicesPre- 1993 n.a. 10 Services Payments due within
12 Domestic goods 3 days of each24 Imported goods transaction.
Proc. 68/1993 17 specific products and all n.a. 5 Services with incomeservice providers income ofbelow Birr 25 daily 5 Birr 25-50 daily
Agricultural and10 essential goods
Services with income12 >Birr 25 daily
All other goods,whether domestic orimported
Amend. 77/1997 17 specific products and all n.a. Increases paymentservice providers income deadline to 7below Birr 25 daily working days from
Reg. 2/1996 169 duty free items 5 minimum Classification of60 maximum items reduced from
5,332 items to 5,294items
Coffee SurtaxPre-1993 Birr 5/quintal cess
Surtax keyed to exportprice
2 levied on exportreceipts less othertaxes
Proc. 67/1993 All pre-1993 leviesReg. 122/1993 plus new levy ofBirr
15/quintalSources: Federal Democratic Government of Ethiopia, selected federal proclamations and the Ministry of Finance,1997, ABC of Taxes in Ethiopia (1942-1996).
4. Institutional reforms. Significant reforms have been initiated, primarily at theFederal level of Government. The revenue collection departments that existed under thecentrally planned economy were structured in accordance with the needs of that centralplanning. With the move to a private oriented market economy a need arose torestructure the workings of the Inland Revenue Authority and Customs, to meet thedemands of a reformulated tax policy. With the Constitution enshrining the right of theState to raise tax revenues, the Inland Revenue Authority personnel were split betweenthe State tax authorities and the Federal revenue authority. On reorganization, theFederal revenue authority was subsequently renamed the Federal Inland RevenueAuthority (FIRA).
22 Appendix 3
IGenera Manager
Management Information Service Internal Audit Service
PR and Customer Service Planning and Tratning Service
Deputy General Manager
Tax Operations Collection and Accounting Branches Legal Administration & Finance
Figure A4.1. Organisation chart of the Federal Inland Revenue Authority
5. The organizational structure of FIRA is shown in Figure A4.1. The FIRA hasfour branch offices in addition to Addis Ababa (Head Office) located at Bahir Dar,Nazret, Awasa, and Dire Dawa. These branch operations are separate from the State TaxBureau (STB). The STB are tax offices at State level, and each region has its ownregional tax office. Where tax due to the Federal Government is difficult, or not costeffective, to collect, an agreement has been reached whereby the STB offices will collectthe tax on behalf of the FIRA in remote areas. Currently this is only done in the Tigrayregion but further cooperation is planned.
6. Prior to the reorganization, the Addis Ababa branch was separately run from theHead Office function. After the ongoing reorganization, the Addis branch will effectivelybecome the FIRA Head office. Although the operational structure has been agreed and isbeing implemented, there is no decision on staff placement at the time of writing. Incommon with other government departments, FIRA is experiencing difficulties in
Ethiopia's Recent Taxation and TariffReforms 23
recruiting new staff due to an inadequate pay and benefits package. About 86-87 percentof the taxation revenue collected by FIRA is collected through the Addis Ababa branch.
7. The Customs administration is currently undergoing a reorganization using therecommendations and outputs from the 1994 report-a modernization and reformprogram. Under this program the main Customs office at La Gare railway station inAddis Ababa, the Addis Ababa Bole International Airport and the customs post atNazeret have been reorganized-the rest is still under way. The Customs Authority Headoffice (in Addis Ababa) has been reorganized and will shortly adopt the followingstructure (Figure A4.2):
| General Manager
l Internal Audit Service l Legal Servicel
|Planning and Statistical Service|
r rDeputy General Manager ||Deputy General Manager
Valuation Customs Duty Free General Cost &L q Police l Priviledges | Services Revenue
Accounts
International Investment & Construction &Relations Duty Drawback M Maintenance
Organisation &L Management
Figure A4.2. Organogram of the Ethiopian Customs Authority
8. The regional customs offices are also organized under the above structure. Themain regional offices are located at Assosa, Dire Dawa (Airport and Railway Station),Moyale, Zala Abnesa, Makelle, Kombolecha and report for management purposes to theDeputy General Manager (Operations). The frontier posts report to the regional offices,except for Moyale which reports directly to the center since there is no Southern regionoffice.
24 Appendix 3
9. Several studies have been carried out in the taxation departments, to identify andrecommend improvements. One major step taken was the formulation of an independentFederal Government Revenue Board (FGRB) to oversee the functions of the two mainFederal departments - FIRA and Customs. 10
Tax Administration Study - Federal Inland Revenue Authority
10. In 1992, the International Monetary Fund (IMF) undertook a study of the reformsrequired to avoid undue risk to the revenue and published a list of recommendations forthe reform of the tax administration. The main points are summarized below:
FIRA Reforms Achievement to DateI Revise tax laws Consolidation reported to be in progress2 Inspection department for regional offices3 PAYE on corporate profits4 Taxation of income not related to any source Dependent on revision of tax laws5 Extend WHT on payments to non-residents6 Set up taxpayers assistance department Planned but not set up7 Enhance collection of tax related information No formal mechanism set up8 Set up tax fraud investigation department No formal mechanism set up9 Train tax auditors Not commenced10 Introduce reward scheme Not commenced11 Start searches for infornation Not commenced12 Start criminal prosecutions for tax fraud Not commenced13 Apply presumptive method of taxation Being applied but at very low rates14 Amend tax clearance certificate procedure15 Prepare instruction manuals Not prepared1 6 Comprehensive computerization Not commenced1 7 Identify and meet staffing requirements Staff requirement survey conducted, but
recommendations not fully met18 Transfer control of all excise goods to Not actioned
Customs
II. In 1995/6. consultants engaged by the United Kingdom Overseas DevelopmentAuthority (ODA now DFID) continued with the reform initiative and building on the IMF work,produced a comprehensive report of the further reforms necessary. This report was subsequentlyadopted by FIRA. On the basis of the report, a further report was prepared by FIRA whichadopted many of the recommendations, and adapted others. This FIRA report was subsequentlyapproved by the Revenue Board, and the implementation of the report's findings is nowunderway. The recommendations in summary were:
FIRA Study recommendationsAssessment and collection
I focus resources on high yielding areas
FGRB also oversees the activities of the National Lottery in Ethiopia.
Fraud control4 set up an investigation unit5 carry out in depth investigations6 authorize monetary settlements as an altemative to prosecution
Finance and accounting7 improve budget setting processes8 adopt double entry bookkeeping9 adopt flexible two line budget10 expand internal audit role to include reviews of controls and management processes
11 Adopt a new organization structure12 Allow FIRA full responsibility for human resources issues including pay structures13 Implement a taxpayer ledger card system14 Implement an information system strategy
Tax Administration Study - Customs Authority
12. The 1992 IMF study mentioned above also published a list of recommendations for thereform of the Customs administration. The main points are summarized below:
Customs Reforms Achievement to DateI Reorganize structure implementation underway2 Selective import declarations not actioned3 Bonus scheme not actioned4 Customs laboratory not actioned5 Finance police to revert to Customs The customs police are under the control of
the management of the CustomsAdministration
6 Regular training programme Actioned but needs further assistance.7 Revise Customs law Actioned8 Customs courts and compounding of Not actioned
offenses9 Streamline clearance procedures New system implemented10 Introduce training Not commenced11 Strengthen verification Not commenced12 Prepare instruction manuals Manuals of instruction prepared13 Adopt ASYCUDA Under implementation14 Exchange information with FIRA Not implemented, except for high level
exchanges15 Modernize Customs warehouse control In process16 Establish Customs zones Not implemented17 Establish anti-smuggling section Enforcement Division in place.
13. In 1994 SGS consultants were engaged to assist the Customs administration to identifyreforms required. Working with five taskforces drawn from within the Customs administration,a number of reforms were recommended in the areas of organization, human resources, control
26 Appendix 3
and classification, valuation, and management information systems. In discussions with themanagement of the Customs administration the following recommendations of the SGS Studyand subsequent implementation status were reported:
SGS Study recommendations Achievement to Date1. Implement a new organization structure A flatter organization with devolved
responsibilities has been agreed and is duefor implementation in March 1998
2. Authority for human resources issues should A job evaluation was carried out and basedbe given to the management of the Customs on the results a pay increase was authorizedadministration at a rate over and above the Civil Service
norm.
The job evaluation presented arecommended increase to the staffingcomplement, which has not yet beenimplemented.
3. Strengthen and simplify clearance and Simplified clearance procedures adopted -control systems reduced to 5 steps.
Manuals compiled.Post-import Control Section formed
4. Strengthen classification methods The World Customs Organization (WCO)Control Notes are used at the Customsstations
5. Strengthen valuation methods GATT procedures adopted.A valuation database is being compiled.
6. Implement a Management Information ASYCUDA system under test at La GareSystem Customs Station.
7. Implement a taxpayer information system All the decrees issued are beingconsolidated for publication
8. Strengthen management Managemenit training is being carried out,with five staff studying for an MBA.Job responsibilities have been agreed andare being put into place.A clieck matrix for management control hasbeen formulated.
APPENDIX 4
GOVERNMENT REVENUE AND FISCALSUSTAINABILITY OF EXPENDITURES IN ETHIOPIA
INTRODUCTION
1. As Ethiopia undertakes reform on all fronts in their recent transition towards amarket-based system and a functioning federation of regional governments, its system ofpublic revenue and structure of government expenditures are also undergoingfundamental changes. To liberalize its economy, for example, the government is seekingto reduce import tariffs significantly. While import tariffs have imparted a pronouncedanti-export bias, they have however contributed as much as a quarter of the tax revenuein Ethiopia'. Thus, the importance of import tariffs in public revenue will likely requirethat trade liberalization be coordinated with macro-fiscal adjustments, namely, acombination of trade-neutral tax increases or public expenditure reduction and a realexchange rate depreciation that is consistent with the liberalization2 .
2. Furthermore, after years of neglect, the new government is also reorienting publicspending towards much needed social investment programs (SIPs) in the areas of basichealth, education, and roads. Its capacity to increase social spending faces severe fiscalconstraints however. For example, the net present value of outstanding public andpublicly guaranteed external debt expressed as a percentage of exports of goods andservices exceeds the critical limit of 250 percent by at least 36 percentage points. Withoutquestion, the economic viability of its strategy for tax reform, increased social spending,and greater economic growth will require debt relief of some kind, like the HIPCinitiative of the World Bank and the IMF. In addition, they depend very much on howEthiopia is able to undertake macro-fiscal adjustment to offset possible revenue lossesfrom trade liberalization and to place its medium-term government expenditure program,particularly in the light of the desired expansion of social spending, in a sustainablesetting. While debt relief is an extremely important issue, its resolutions require joint andagreed actions among the Government, multilateral institutions, and donors, which willnot be the focus in this paper3 . This paper will instead examine the two latter issues: (1)
In comparison, India, which used to have one of the highest protective rates of protection, collected 24percent of tax revenue from import tariffs prior to the trade liberalization it undertook in the nineties.Ethiopia's collection rates from import tariffs are however much lower than India's 60 percent.
2See, for example, Greenaway and Milner (1991), Branson (1992), and Mitra (1992) for a detaileddiscussion of this issue.
3However, it will be looked at from the context of the reductions in debt service payments and theirimplications on the level of sustainable fiscal deficits.
28 Appendix 4
the revenue consequences of tariff liberalization, and (2) the sustainability of themedium-term public expenditure program.
THE FRAMEWORK
3. One feature that characterizes transitional economies like Ethiopia is the lack oftime-series data that are consistent with the emerging market-oriented economy. For thisreason, this paper relies on a simple general equilibrium model of an open economy,based on a modified Salter-Swan framework with imperfect substitution between foreignand domestic goods, to capture the salient aspects of fiscal policy and revenueconstraints. The framework is the 1-2-3 model described in Devarajan, Go, Lewis,Robinson, and Sinko (1997), which can be implemented by making use of little morethan basic national accounts and available revenue information. The goal is to provideillustrative rather than precise estimates of the economy-wide consequences of policyreform and the government's expenditure program. Better estimates are possible as dataimprove, which also makes plausible the construction of more elaborate models and theeconometric estimations of their parameters. Nevertheless, the 1-2-3 model, while small,captures features characteristic of developing countries like Ethiopia and a rich array oftrade and fiscal issues. A major benefit from its simplicity is the increased transparencyof the mechanisms by which a policy reform or an external shock affects the economy.Moreover, the model behaves in a similar fashion to more complex multisector models,so that it anticipates the results that can be obtained from the multisector models4
4. In this paper, the framework is also extended to look at the fiscal sustainability ofthe government expenditures and growth over the medium term in relation to anticipatedresources in Ethiopia. In simple terms, sustainability becomes an issue when anexpansionary fiscal policy occurs in a situation when the real interest rate is above thereal growth rate of an economy so that the increase in public debt will feed upon itselfand the debt level eventually become excessive in relation to exports and othermacroeconomic variables. Fiscal sustainability is therefore better viewed over time, withan explicit or implicit examination of the government's intertemporal budget. Anoverexpansion of the fiscal budget must involve either a contractionary fiscal policy inthe future (Ricardian equivalence) or greater inflation from an increase in seignoragefrom money creation. In Ethiopia, where foreign financing and external debt reschedulingand relief are critical factors, fiscal and external debt sustainability are closely linked.
5While it is feasible to employ a fully dynamic, intertemporal version of the 1-2-3 model5,a more practical, recursive (year-to-year) time-series version is adopted, focusing more onthe effects of the revenue and expenditure plans on the primary deficit of the public sectorand its financing options over the medium term.
4 Recent examples of very disaggregated CGE models include the 72-sector model for household groupsby Devarajan and Muensburger (1997)
5 See Devarajan and Go (1998).
Government Revenue and Fiscal Sustainabilitv of Expenditures in Ethiopia 29
Description of the Model
5. The list of equations in the model is presented in Table 1. On the production side,a production possibility frontier determines the efficient combinations of E and D that theeconomy can supply. This frontier is specified as a constant elasticity of transformation(CET) function with transformation elasticiity Q (Equation 1), or more specifically,
X = A, [ E PI + (I 6, )D P,
here Q =1/(p,- 1) and 1 <p, < + c. Aggregate output (X) is fixed, equivalent to
assuming full employment of factor inputs. To maximize profits, producers choose theoptimal ratio of exports to domestic goods (E/D) based on their relative prices (Equation2), i.e.,
E _(I-6 _ P_D L .p
Equation 3 arises from the homogeneity assumption or zero-profit condition.
6. On the demand side, the consumer purchases a composite commodity made up ofdomestic good D and imports M (Equatiorn 4). Imports and domestic goods are assumedto be imperfect substitutes so that the composite good is defined by a constant elasticityof substitution (CES) aggregation function of M and D, with substitution elasticity6 (Equation 4), or more specifically6
Q = Aq[6 qMAJ' +(I -6 q)Dp" q
The consumer maximizes utility or welfare by choosing an optimal ratio of M to D basedon their relative prices (Equation 5), which takes the following specifi- form:
D (I - 6 q)Pw,
where a = 1(1 - Pq) and - x < Pq < +C . In any system of expenditure equations, the
value of purchases of domestic goods and imports must equal total expenditure (Equation6). Total demand for the composite good is an aggregation of household consumption,
6 See Armington (1969).
30 Appendix 4
government consumption, and total investment (Equation 7); the determination of each ofthese final demand components is discussed below.
7. The household sector receives total income generated from production plus nettransfers and net factor income from the government and the rest of the world (Equation8). The household then saves a fixed proportion of its income and spends the rest onconsumption (Equation 9).
8. On the fiscal accounts, the government collects four broad categories of taxes: animport tariff t,,,, an export duty or subsidy t, an indirect tax on domestic sales, t, and a
direct tax rate t, (Equation 10). Public savings (budgetary deficit or surplus from current
operations) is the balance of tax revenue plus foreign grants and government expenditures(all exogenous) such as government consumption and transfers to households (Equation11).
9. In Table 1, the price equations define relationships among prices andcorresponding tax wedges. There are fixed world prices for E and M; domestic prices forE and M; the price of the domestic good D; and prices for the two compositecommodities, X and Q. The domestic prices of exports and imports include the pricewedges created by trade taxes or subsidies (Equations 12 and 13). The sales price of thecomposite good includes a sales tax (Equation 14). The expenditure accounting identitiesin Equation 3 and 6 defines the price of output and the composite good, respectively7 .Note that the system is homogeneous of degree zero in prices -- doubling all prices, forexample, leaves real demand and the desired export and import ratios unchanged. Sinceonly relative prices matter, it is necessary to define a numeraire price; in equation 15, thisis specified to be the nominal exchange rate, R in the basic setup.
10. Market-clearing equilibrium requires that: supply must equal demand for D and Q(Equation 16 and 17); the balance of trade constraint must be satisfied (Equation 18); and,total savings must equal the cost of investment (Equation 19). Equation 20 defines thesources of total savings. The complete model has 20 equations and 19 endogenousvariables. The equilibrium conditions, however, are not all independent by Walras' Law.Any one of them can be dropped and the resulting model is fully determined. models.
Trade Liberalization, Macro Balances, and Fiscal Adjustment
11. There are several modes of adjustment possible to a policy reform like tradeliberalization (or to an adverse external shock). In an economy like Ethiopia, theavailability of foreign exchange or financing is assumed to be the binding resourceconstraint. An external balance in the economy is attained when a prescribed value of the
The dual price equations for output and the composite good can also be used. However, in practice,it isoften convenient to use expenditure identities, invoking Euler's theorem for linearly homogeneousfunctions.
Government Revenue and Fiscal Sustainabilitv of Expenditures in Ethiopia 31
current account deficit in the balance-of-payments is achieved or maintained. The focusof the analysis is to bring about such external balance through a reduction of domesticabsorption using fiscal adjustment as policy instrument. The first type of adjustment is toraise taxes, mainly through an important revenue source and trade-neutral taxation (i.e.,domestic indirect taxes like a sales tax or VAT that applies to both domestic goods andimports) while keeping government expenditure constant in real terms. Since with fixedexpenditures, the government saves all additional income (whereas the private sectorsaves only part of its additional income), domestic savings is increased by transferringincome to the public sector, i.e., by increasing domestic indirect taxes. In the second typeof fiscal adjustment, the government reduces domestic demand by cutting its own non-investment expenditures. In either case, domestic savings are raised to finance thedifference between total investment and the exogenously specified current account deficitof the balance-of-payments. While the focus is on tax adjustment, the expenditureadjustment is also presented as an option. In particular, the potential loss in Tax revenue
T from a tariff reduction t,,(with no other tax changes, i.e., t,=O,t, = 0) may be
expressed as
I 0 'R (t+M)+H,R ( )+R!
where a carat over a variable is the log differentiation of the variable, denoting a small
percentage change. 0, R' is the relative weight of the tax revenue from source i and,R
hence, S0,'j = I . It can be shown that the revenue loss is equal to
T = k(o ,Q,)t,,
k(o ,Q,...) is generally a positive constant, which depends on the elasticities and otherparameters in the model so that tariff reform implies a revenue loss8 . Tax adjustmentrefers to a change in the domestic indirect tax, t,, so that the second term above, now
reflecting the compensating tax adjustment. 0,qR(i, +P, + Q), makes T =0. Expenditure
adjustment refers to a change in real government consumption G so that (P, + G) = T.
12. Moreover, the absorption contraction caused by the fiscal adjustment (tax increaseor expenditure reduction) undertaken will put downward pressure on domestic prices,causing a real depreciation of the exchange rate. This exchange rate behavior is akin tothe expenditure-switching policy necessary to remove or reduce an international tradeimbalance. As in standard macroeconomic theory, such depreciation will cause exports toexpand in order to offset the domestic contraction caused by the fiscal adjustment, such
8 See Devarajan, Go, and Li (1998) for a derivation of reduced forms for various tax structures.
32 A=endix 4
that output can be maintained at full (or base-year) employment. It will also help curtailthe expansion of imports (i.e., prevent it from becoming too cheap from the tariffreduction) in a manner consistent with the targeted macro balance. In the 1-2-3 model, thereal exchange rate is the relative price ratio of foreign (e.g. imports) and domestic goods,RER = pw,, /Pd. Given that the world price of imports is fixed,
RER=-PlDomestic inflation (deflation) leads to a real appreciation (depreciation) in the exchange
rate. It can be shown that Pd varies directly with i,:9
Where o ,,,, = is the price wedge created by t,,, .
Government Expenditures, Medium-term Growth, and Fiscal Sustainability
13. The fiscal sustainability of government expenditures in Ethiopia, especially inanticipation to the possible revenue loss from the trade reform and the expanded socialexpenditures, depend critically on four key resources over the medium term: donor orofficial grants to the government, debt relief and change in arrears based on the ParisClub aggreements, additional domestic sources of finance that do not add to governmentdebt, and possible additional borrowing based on concessionary terms like IDA loans.The first two, donor money and debt relief are exogenous while the latter two are withingovernment's policy. As long as domestic financing does not add to public debt andforeign financing is concessionary with 10 or more years of grace period, the real cost ofadditional resources will not exceed the real growth of the economy, which withcontinuing liberalization and privatization will grow with continuing efficiency, thusensuring its sustainability. An additional option is to increase the number of yearsrequired to implement the expanded social expenditures and lowering its annual cost.
14. To examine these various financing options, a time-series version of the model isemployed. For each period, the following balance-of-payment constraint is employed inplace of the current account balance, Equation 18, in Table 1. The BOP financing gapgap is the balance of trade of goods and services net of various projected (exogenous)dollar inflows. These foreign exchange inflows include net factor income to the privatesector, net factor income of the government nfty, net foreign transfers to the private
sector nftpI, net official transfers or grants to the government nftg, net short-term and
long-term capital inflows ncap, IMF credit and change in other foreign liabilities in the
Ibid.
Government Revenue and Fiscal Sustainabilit of Expenditures in Ethiopia 33
Central Bank account ofin change in the official dollar reserve reserv (endogenouslyequivalent to four month's imports), and very important in Ethiopia's medium-termfuture, the expected change in arrears and debt relief relief.
gap = pw, M - pw1.E
-nfyp -nfyg
-f 7fp - nftg
- ncap - ofin
- reserv - relief
15. We also expand the government budget to include various projected financingflows. The overall fiscal balance is the projected current and capital public expenditures,P, (G + Zg), tax revenue Tax (including new taxes), net total transfers to the privatesector (including transfers, subsidies, interest payments on domestic debt, and nontaxrevenue), donor money in birr nftg.R, privatization proceeds priv.P,, and additionalforeign financing gap.R.
Ibal = P, (G + Zg) - Tax
-P,qtrns - nftg.R
- relief.R - priv.Pl1
- gap.R
34 Appendix 4
Table 1: The 1-2-3 Model with Government and Investment
Production Government Budget(1) X = G(E,D; Q) (10) T = tm R pwm M + tS pq QD
(2) E/D5 =g 2 (Pe,P) + tY y +t' R pw' E(3) PXX = PeE+PdDs ~(11) T-PIG-trP'+ftR-S';=o
(5) M/DD f,(pm,pd)(I)Pe=I tRpw(6) pq QS pin M + P'DD (1 4) P= (I + t') RPw
(7) QD=C+Z+ G (15) R=I
Houselsold Budget Eauilibrium Conditions(8) Y=PXX + tr + reR (16) D0 -Ds=O(9) c P' = (Il - - ty) Y ( 17) QD _QS = 0
(18) pwnM-pweE-ft-re= B(I 9) P t Z - s = 0
(20) S= s Y+RB +Sg
Endogenous Variables: Evogenous Variables:
E: Export good pw'": World price of import goodM: Import good pwe: World price of export goodDS: Supply of domestic good t"': Tariff rateDD: Demand for domestic good tc: Export subsidy rateQS: Supply of composite good ts: sales/excise/value-added tax rateQD: Demand for composite good ty: direct tax ratePe: Domestic price of export good tr: government transfersPm: Domestic price of import good ft: foreign transfers to governmentpd: Producer price of domestic good re: foreign remittances to private sectorP': Sales price of composite good .s:Average savings ratepX: Price of aggregate output x: Aggregate outputPI: Price of composite good G: Real government demandR: Exchange rate :T: Tax revenue B: Balance of tradeS': Government savings Q: Export transformation elasticityY: Total income a : Import substitution elasticityC: Aggregate consumptionS: Aggregate savingsZ: Aggregate real investment
APPENDIX 5
STATISTICAL TABLES
Table 1. Recurrent Expenditure by Functional Classifications from 1986/87-1992/93Table 1. Recurrent Expenditure by Functional Classifications from 1993/94-1994/95Table 1. Recurrent Expenditure by Functional Classifications from 1995/96-1997/98Table l.a. Recurrent Expenditure by Functional Classifications from 1986/87-1992/93 as a % of
Total ExpenditureTable l.a. Recurrent Expenditure by Functional Classifications from 1993/94-1994/995 as a % of
Total ExpenditureTable l.a. Recurrent Expenditure by Functional Classifications from 1995/96-1997/98 as a % of
Total ExpenditureTable 2. Capital Expenditure by Functional Classifications from 1986/87-1992/93Table 2. Capital Expenditure by Functional Classifications from 1993/94-1994/95Table 2. Capital Expenditure by Functional Classifications from 1995/96-1997/98Table 2.a. Capital Expenditure by Functional Classifications from 1986/87-1992/93 as a % of
Total ExpenditureTable 2.a. Capital Expenditure by Functional Classifications from 1993/94-1994/95 as a % of
Total ExpenditureTable 2.a. Capital Expenditure by Functional Classifications from 1995/96-1997/98 as a % of
Total ExpenditureTable 3. Total Gov. Rec. Expenditure by Economic Classifications from 1986/87-1992/93Table 3. Total Gov. Rec. Expenditure by Economic Classifications from 1993/94-1994/95Table 3. Total Gov. Rec. Expenditure by Economic Classifications from 1995/96-1997/98Table 3.a. Total Gov. Rec. Expenditure by Economic Classifications from 1986/87-1992/93 as a
% of Total ExpenditureTable 3.a. Total Gov. Rec. Expenditure by Economic Classifications from 1993/94-1994/95 as a
% of Total ExpenditureTable 3.a. Total Gov. Rec. Expenditure by Economic Classifications from 1995/96-1997/98 as a
% of Total ExpenditureTable 4. Education Recurrent Expenditure by Economic Classifications from 1986/87-1992/93Table 4. Education Recurrent Expenditure by Economic Classifications from 1993/94-1994/95Table 4. Education Recurrent Expenditure by Economic Classifications from 1995/96-1997/98Table 4.a. Education Recurrent Expenditure by Economic Classifications from 1986/87-1992/93
as a % Total ExpenditureTable 4.a. Education Recurrent Expenditure by Economic Classifications from 1993/94-1994/95
as a % Total Expenditure
36 Avpendix 5
Table 4.a. Education Recurrent Expenditure by Economic Classifications from 1995/96-1997/98as a % of Total Expenditure
Table 5. Primary Education Recurrent Expenditure by Economic Classifications 1986/87-1992/93
Table 5. Primary Education Recurrent Expenditure by Economic Classifications 1993/94-1994/95
Table 5. Primary Education Recurrent Budget by Economic Classifications 1995/96-1997/98Table 5.a. Primary Education Recurrent Expenditure by Economic Classifications 1986/87-
1992/93 as a % of Total ExpenditureTable 5.a. Primary Education Recurrent Expenditure by Economic Classifications 1993/94-
1994/95 as a % of Total ExpenditureTable 5.a. Primary Education Recurrent Budget by Economic Classifications 1995/96-1997/98 as
a % of Total ExpenditureTable 6. Health Recurrent Expenditure by Economic Classifications from 1986/87-1992/93Table 6. Health Recurrent Expenditure by Economic Classifications from 1993/94-1994/95Table 6. Health Recurrent Expenditure by Economic Classifications from 1995/96-1997/98Table 6.a. Health Recurrent Expenditure by Economic Classifications from 1986/87-1992/93 as a
% of Total ExpenditureTable 6.a. Health Recurrent Expenditure by Economic Classifications from 1993/94-1994/95 as a
% of Total ExpenditureTable 6.a Health Recurrent Budget by Economic Classifications from 1995/96-1997/98 as a %
Total ExpenditureTable 7. Primary Health Recurrent Expenditure by Economic Classifications 1986/87-1992/93Table 7. Primary Health Recurrent Expenditure by Economic Classifications 1993/94-1994/95Table 7. Primary Health Recurrent Expenditure by Economic Classifications 1995/96-1997/98Table 7.a. Primary Health Recurrent Expenditure by Economic Classifications from 1986/87-
1992/93 as a % of Total ExpenditureTable 7.a. Primary Health Recurrent Expenditure by Economic Classifications from 1993/94-
1994/95 as a % of Total ExpenditureTable 7.a. Primary Health Recurrent Expenditure by Economic Classifications from 1995/96-
1997/98 as a % of Total ExpenditureTable 8. Agriculture Recurrent Expenditure by Economic Classifications 1986/87-1992/93Table 8. Agriculture Recurrent Expenditure by Economic Classifications 1993/94-1994/95Table 8. Agriculture Recurrent Expenditure by Economic Classifications 1995/96-1997/98Table 8.a. Agriculture Recurrent Expenditure by Economic Classifications from 1986/87-1992/93
as a % of Total ExpenditureTable 8.a. Agriculture Recurrent Expenditure by Economic Classifications from 1993/94-1994/95
as a % of Total ExpenditureTable 8.a. Agriculture Recurrent Expenditure by Economic Classifications from 1995/96-1997/98
as a % of Total ExpenditureTable 9. Transport Constructions Recurrent Expenditure by Economic Classifications 1986/87-
1992/93
Statistical Tables 37
Table 9. Transport Constructions Recurrent Expenditure by Economic Classifications 1993/94-1994/95
Table 9. Transport Constructions Recurrent Expenditure by Economic Classifications 1995/96-1997/98
Table 9.a. Transport Constructions Recurrent Expenditure by Economic Classifications from1986/87-1992/93 as a % of Total Expenditure
Table 9.a. Transport Constructions Recurrent Expenditure by Economic Classifications from1993/94-1994/95 as a % of Total Expenditure
Table 9.a. Transport Constructions Recurrent Budget by Economic Classifications from 1995/96-1997/98 as a % of Total Expenditure
Table 10. Capital Expenditure by Economic Classifications from 1986/87-1997/98Table 11. Education Capital Expenditure by Economic Classifications from 1986/87-1997/98Table 12. Health Capital Expenditure by Economic Classifications from 1986/87-1997/98Table 13. Agriculture Capital Expenditure by Economic Classifications from 1986/87-1997/98Table 14. Transport Constructions Capital Expenditure by Economic Classifications from
1986/87-1997/98Table 15. Federal and Regional Governments Recurrent Actual Expenditure for 1993/94Table 16. Federal and Regional Governments Recurrent Actual Expenditure for 1994/95Table 17. Federal and Regional Governments Recurrent Actual Expenditure for 1995/96Table 18. Federal and Regional Governments Recurrent Preliminary Expenditure for 1996/97Table 19. Federal and Regional Governments Recurrent Budget for 1997/98Table 20. Federal and Regional Governments Capital Actual Expenditure for 1993/94Table 21. Federal and Regional Governments Capital Actual Expenditure for 1994/95Table 22. Federal and Regional Governments Capital Actual Expenditure for 1995/96Table 23. Federal and Regional Governments Capital Preliminary Expenditure for 1996/97Table 24. Federal and Regional Governments Capital Budget for 1997/98Table 25. FY 1993/94 Regional Governments Revenue, Expenditure and Budget TransfersTable 26. FY 1994/95 Regional Governments Revenue, Expenditure and Budget TransfersTable 27. FY 1995/96 Regional Governments Revenue, Expenditure and Budget TransfersTable 28. FY 1996/97 Regional Governments Revenue, Expenditure and Budget TransfersTable 29. FY 1997/98 Regional Governments Revenue, Expenditure and Budget TransfersTable 30. Capital Expenditure by Sources of Finance from 1986/87-1997/98Table 31. Government Revenue Performance from 1986/87-1997/98Table 32. Federal and Regional Governments Revenue for 1993/94Table 33. Federal and Regional Governments Revenue for 1994/95Table 34. Federal and Regional Governments Revenue for 1995/96Table 35. Federal and Regional Governments Preliminary Revenue Estimate for 1996/97Table 36. Federal and Regional Governments Revenue Estimate for 1997/98Table 37. Ethiopia: Selected Macroeconomic Indicators from 1984-1998
38 Appendix 5
CODE RECURRENT EXPENDITURE CODE DESCRIPTIONS
6100 PERSONAL SERVICES6101 Civil Salaries6102 Civil Allowances6103 Military and Police Pay6104 Military and Police Allowances6202 Traveling, Subsistence and Transportation
6200 NON-PERSONAL SERVICES6201 Public Utilities Services6203 Printing and Advertising6204 Repairs and Maintenance of Premises and Equipment6205 Repairs and Maintenance of Motor Vehicles6206 Rentals for Building and Equipment6207 Tax6210 Miscellaneous Contractual Services
-5300 MATERIALS AND SUPPLIES6301 Food Stuffs6302 Medical Supplies and Equipment6303 Educational Supplies6304 Uniforms, Clothing, Bedding6305 Petrol and Lubricants6306 Office Supplies6307 Other Materials and Supplies
6400 GRANTS, CONTRIBUTIONS AND OTHER CURRENT TRANSFERS6401 Grants to Individuals6402 Grants to Institutions6403 Contribution to International organization6411 Civil Pension Payment6412 Military Pension Payment6413 Daregot /Supplementary Pension Payment/6425 Price Subsidy6431 Retained Revenue6451 External Debt Principal6453 External Debt Interest Payment6457 Domestic Debt Principal6458 Domestic Debt Interest Payment
6500 PURCHASE OF MOTOR VEHICLES AND EQUIPMENT6501 Purchase of Motor Vehicles
Statistical Tables 39
6502 Purchase of Equipment
6600 MILITARY CONSTRUCTIONS AND EQUIPMENT6601 Military Construction and Equipment6602 Arms and Ammunition
CODE CAPITAL EXPENDITURE CODE DESCRIPTIONS
8101 Feasibility and Other Studies8102 Engineering Designs8201 Residential Buildings8202 Non-Residential Buildings8203 Land Preparation and Other Constructions (Roads, Dams, Bridges, etc.)8204 Machinery and Equipment (Bulldozers, Tractors etc.)8205 Transport Machinery (Motor Vehicles)8206 Office Equipment8207 Livestock8301 Project Management and Supervision8302 Direct Labor Cost8303 Project Operating Costs8304 Financial Costs (Interest, Commission, Insurance & Others)8305 Taxes8401 Capital Contribution8402 Debt and Advance Payments8403 Capital Grants8404 Compensation Payments8000 To be allocated when information is available
REGIONS
Region 1 TigrayRegion 2 AfarRegion 3 AmharaRegion 4 OromiyaRegion 5 SomaliRegion 6 Benshangul-GumuzRegion 7-11 South Nation Nationalities PeopleRegion 12 GambellaRegion 13 HarariRegion 14 Addis AbabaRegion 15 Dire Dawa
Table 1. Recurrent Expenditure by Functional Classifications from 1986/87-1992/93In Million Birr
86/87 87188 88/89 89/90 90/91 91/92 92193 92193
ACTUAL ACTUAL ACTUAL ACTUAL ACTUAL ACTUAL REVISED ACTUAL
Total 2,620 3,420 3,786 3,643 3,636 3,254 3,43sAdjustments have been made In order to much the total of economic classlfications with the total of functional classifications.
Table 3. Total Government Recurrent Expenditure By Economic Classifications from 1993/94 - 1994/95In Million Birr
1993/94 1993/94 1994/95 1994/95
code Budget Actual Budget Actual
Region Center National Region Center National Region Center National Region Center National
Total 2,293 2,956 5,249 2,471 3,112 5,583 2,525 3,130 5,654 2,909 3,197 6,106^ Adjustments have been made in order to much the total of economic classlfications with the total of functional classifications.
Table 3.a. Total Government Recurrent Expenditure By Economic Classifications from 1986/87-1992/93As a Share of Total Expenditure
code Actual Actual Actual Actual Actual Actual Actual
6100 95.9% 96.4% 97.0% 97.7% 97.8% 97.3% 97.3%
6101 94.8% 95.4% 96.0% 96.8% 97.2% 96.5% 96.8%
6102 0.6% 0.6% 0.6% 0.6% 0.4% 0.5% 0.3%
6103 0.0% 0.0% 0.0% 0.0% 0.0% 0.1% 0.0%
6104 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6202 0.5% 0.4% 0.4% 0.3% 0.2% 0.2% 0.2%
6200 0.7% 0.5% 0.4% 0.4% 0.2% 0.3% 0.2%
6201 0.2% 0.1% 0.1% 0.1% 0.1% 0.1% 0.1%
6203 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6204 0.4% 03% 0.2% 0.2% 0.1% 0.1% 0.1%
6205 0.0% 0.0% 0.0% 0.0% 0,0% 0.0% 0.0%
6206 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6207 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6210 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6300 0.8% 0.8% 0.8% 0.8% 0.6% 0.7% 0.7%
6301 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6302 0.0% 0.0% 0.0% 0.0% 0.0% 0 0% 0.0%
6303 0.3% 0.2% 0.3% 0.3% 0.2% 0.2% 0.2%
6304 0.3% 0.3% 0.3% 0.3% 0.3% 0 3% 0.3%
6305 0.1% 0.1% 0.0% 0.1% 0.0% 0.0% 0.0%
6306 02% 0.1% 0.1% 0.1% 0.1% 01% 0.1%
6307 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0,0%
6310 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6400 2.6% 2.2% 1.8% 1.0% 1.4% 1.7% 1.7%
6401 0.1% 0.0% 0.0% 0.0% 0.0% 0.e% 0°0%
6402 2.5% 2.2% 0.2% 1.0% 1.4% 1.7% 1.7%
6403 0.0% 0.0% 1.6% 0.0% 0.0% 0.0% 0.0%
6411 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6412 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6413 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6425 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6431 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6451 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6453 00% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6457 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6458 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6500 0.1% 0.1% 0.0% 0.1% 0.0% 0.0% 0.0%
6501 0 .0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%
6502 0.1% 0.1% 0.0% 0.1% 0.0% 0.0% 0.0%
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%- 100.0%
Table 5.a. Primary Education Recurrent Expenditure by Economic Classifications from 1993/94 - 1994/95As a Share of Total Expenditure
1993194 1993194 1994195 1994195
code Budget Actual Budget Actual
Region Center Nabonal Region Center Nabonal Region Center National Region Center National6100 91.2% 63.4% 91.1% 91.6% 56.7% S1.4% 91.2%/ 53.4% 91.1% 92.0% 73.6% 911.9%
Total 304.58 56.86 361.44 270.86 57.20 328.06 302.08 61.03 363.11 335.71 62.08 397.79* Adjustments have been made in order to much the total of economic dassmfications with the total of functional classificatons.
TIble6.a Health Recurrent Expenditure by Economic Classifications from 1986/87 - 1992/93As a Share of Total Expenditure
Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%*Adjustments have been made in order to murh the total of economic dassificafions with the total of functonal classificabons.
Table 6.a. Health Recurrent Expenditure by Economic Classifications from 1995/96 - 1997/98As a Share of Total Expenditure
1995/96 1995/96 1996/97 1997/98
code Budget Actual Budget Budget _ _
Region Center National Region Center Nabonal Regions Center National Region Center Nabonat
* Federal and Regional States recurrent expendilture does not icnude extemal assistance
1/ The Federal Slate preliminary expenditure for Natural Resources is induded under Agriculture
TABLE 19. FEDERAL GOVERNMENT AND REGIONAL GOVERNMENTS RECURREN T BUDGET FOR FY97/98In Million Birr
Federal Tigray Afar Amhara Oromiya Somali Benshangul SNNP Gambella Haran Addis Ababa D'Dawa Regions NabonalDescriptlons State 1 2 3 4 5 6 7-11 12 13 14 15 Total Total
Table 30. Capital Expenditure by Sources of Finance
In Million Birr _
1986/87 1987/88 1988/89 1989/90 1990/91 1991/92 1992/93 1993/94 1994/95 1995/96 1996/97 1997/98 1997/98Actual Actual Actual Actual Actual Actual Actual Actual Actual Actual Pre.Actual Budget Pre.est.
Million Ethiopian Birr Actual Pre, Actual Estimate Budget.............................................................. .................. ................... --------.... ._.._
GROSS DOMESTIC PRODUCT
GDP AT MARKET PRICES 10,063 12,165 12,617 13,385 13,869 14,631 17,872 19,195 20,792 26,671 28,329 33,885 37,938 41,465 45,204