DeUmett of The World Bank FOR OQIICIAL USK ONLY k ,a,~~~~~~1-4 ae AZ. ahpen No. 7093-AR STAFF APPRAISAL REPORT ARGENTINA MUNICIPAL DEVELOPMENT PROJECT February 26, 1988 Infrastructure and Energy Operations Division Country Department 4 Latin America and the Caribbean Region Tids docmmet as a r1ested db.tim and my he used by reciW omy in the perfomsame of tbeir odal dutes. Its _ntent my n_ oterws be disosed _ .o d Wodank auc ho mon. Currency Unit - Austral (A) On October 14, 1987, * two-tiered exchange rate syst _ vas announced: Co=ercial Fixed Rate (official): US$1.00 - A 3.50 A 1.00 - US$0.29 Free Market Rate (Appraisal: Novembr 14, 1987)s US$1.00 - A 4.03 A 1.00 - US$0.25 Fiscal Year January 1 - December 31 WEIGHTS AND MEASURES The Metric System has been used throughout this report. ABBREVIATIONS AMBA Buenos Aires Metropolitan Area (Area Metropolitana de Buenos Aires) FONAVI National Housing Fund (Fondo Nacional de la Vivienda) INDEC National Census Bureau (Instituto Nacional de Estadistica y Censos) PEU Project Execution Unit POM Project Operations Manual SVOA Secretariat of Housing and Environmental Management (Secretaria de Vivienda y Ordenamiento Ambiental) FM3 OFflICA UU ONdLY I ~~~~MVICIPZL DIYtLPUl" MMOI table of Contoent upb. LOWM CS 8 SUIIART ................................. iLI I. THE I Recent Urbanisotion Trends............................... 1 Oovernaet Structure for Provision of Public S.rvic...... 2 Public Sector Financing and Intergovernmental Fiscal Relations .................................... 2 Sector Development Issues and Obj*ctives................. 5 Sector Lending Strategy and Rationale for Bank Involv _ nt ................................. 5 II. THE ........................................... ........ 7 Project Origin .................................. ......... 7 Project Objectives ....................................... 7 Project Components....................................... 7 Criteria for Implementation of Project Objectives........ 7 Project Costs and Allocation by Province................. 11 Financing Plan ............................ 13 Implementation Arrangements 5.......... 15 P r o c u r e m e a t ~~~~~~~~~~~~16 Di bur empat ~~~~~~~~~~~18 Accounts and A uAs19 conomic ....Impa.......................................... 21 Poverty Impt............ .. . . 21 Environmental Impact............ 22 Risks and Safeguards ..................................... 22 III. AGRE3MNTS REACHED ANDC REC O4NDATIONSD... TI..O........ 23 This report Is based on the findings of a pre-appraisal mission that visited Argentina from August 10 to September 3, 1987. This mission was comprised of Meosrs. James Hicks (Mission Loader, LAI4E), Mario Artasa- Rousel (LA4IE), David Vettor and Lubomir Ficicski Dunin (Consultants). The appraisal mission yes undertaken by Mr. Hicks during November 11-16, 1987. Mrs. Marli Nikolov-Fernandes assisted in the production of this report. Thistdocumnths astdrcted drbution ndouYbou byrspbonlyIn dPfonn of thdkofioWdudueL Its contents muynot oftnvinbebdisckFo witout Wodd Ban nsuhodudm - ii - Table of Contents (Continued) LIST OF AMINES P nas Mo. 1. Oeneral Demographic and Economic Trends and Characteristics of Participating Provinces.... 26 A. Demographic Treds. ................................... 26 3. Economic Aspects .... 28 C. Int-r-Governmental Transfers and the Capacity to Plan Capital Investments at the Municipal Level... 29 2. Municipal Investment Plan Prof41 s 52 A. Physical Investments... 52 B. Technical Assistance and Triining 53 3. Implementation Arrangementsr...... 56 4. Project Impact on Provincial and Municipal Financess.... * 60 5. Project Economic and Social Analysis..................... 64 6. Provincial Project Operations Ma as .ls 72 7. Major Cost Tables 1. Typical Investment Component Unit Costs .............. 74 2. Project Costs by Component........................... 75 3. Detailed Project Costs and Contingencies............. 76 4. Investment Schedule by Major Component............... 77 5. Estimated Schedule of Disbursements.................. 78 i 8. Selected Documents and Data Available in Project File.... 79 MAP IBRD 20486 -- Argentina: Municipal Development Project ARGENTINA MUNICIPAL DEVELOPMENT PROJECT LOAN AND PROJECT SMtSKY Argentine Republic. Provinces of Bueno Aires, Cordoba, La Papa, Neuquen, and Santa Fat and eligible municipalities of tbese province. Amount: US$ 120.0 million equivalent. Terms: Repayment in 15 yes"s, including three years of grace, with interest at the Bank's standard variable rate. Relendina T rms: All proceeds of the loan would be onlent from the Central to Provincial Governments under the same terms and conditions as the Bank loan, except that the Central Government would assume the cross- currency exchange risk and pass on the US dollar xchange risk to provinces, with the condition that if the periodic, dollar-indexed debt service adjustments exceed a band' of plus or minus 102 deviation from the official inflation rate, then the difference would be credited to (or capitalixed in) outstanding provincial debt. In addition, provinces would pay a service charge of up to 12 of loan amounts disbursed. Except for on amount up to 51 of the loan (or 101 in exceptional cases) that may be used by the provinces to finance technical assistance and training, all proceeds of the Bank loan would be onlent from provinces to municipalities on harder" terms (maxim4m of 10 years repayment, including up to one year of grace, with spread of at least 1.5 percentage points over interest rate paid by provinces to Central Government) necessary to capitalize a Municipal Development Fund for project replication in each province. Municipal repayment of subloans would be guaranteed through the province's retention of the mmnicipality's debt servicing obligations from provincial revenue sharing. Prolect Obiectives and Description: The main purpose of the proposed project is to contribute to more effective public sector management at the provincial and municipal levels through improved financing mechanisms for municipal investments. The specific objectives of the project would be to: (a) mobilize external and internal - iv - resources in a non-deficit and non-inflationary wsy to finance justified municipal investmentsg (b) strengthen municipalities' capacity to plan, finance and execute cost-effective capitol mprovcmnts progr_me and (c) promote structured, periodic consultations between municipal and provincial authorities for the formulation and evaluation of Investment plans. The project would provide subloans to the five bcnefitting provinces for financing SO1 of total eligible cost* fort (a) physical Investments, including construction and rehabilitation of public infrastructure and coinmity facilities, and purchase of vehicles and equipments and (b) technical assistanco and training. In order to implement effectively the project's objectives, only creditworthy municipalities (having a balanced current account and debt obligations within acceptable limits) would be eligible to receive subloans for physical investments, and each annual investment plan of each eligible municipality must have at least 65. of total proposed physical investments allocated to components for which 1002 of total investment costs will be fully recovered from payments to the municipality by direct beneficiaries (betterment levies, connection fees, etc.). Benefits: There are four main categories of benefits: (a) financial, where the project's design criteria (primarily through limitation of onlending to creditworthy municipalities that agree to recover costs substantially in excess of the loan amount) would assure that incremental resources mobilized through the project would be used in a non-deficit and non-inflationary way, and municipalities that participate in the project would be financially strengthened to provide other services not financed under the project; (b) poverty Imact, with long- term financing making cost recovery affordable to the poor, and with creation of approximately 58 thousand person/years of construction work; (c) institutional, through providing financial predictability necessary for sub-national governments to plan and program investments; and (d) economic, with an average IRR of 382 for infrastructure investments. In addition, it is anticipated that the lessons learned from the design and implementation of this operation will make the project's replication in other provinces relatively easy. Risks and Safetuards: Possible risks include ineffective provincial management and lack of continuity in project support. The main safeguard for this would be to spread the risk among the five provinces. This imuld be **meve primzily throuh the o ditie that It im"matatia delays ownc la province. them the liitial losm ellocatdo. by pr.viae ~msl be adjusted to allocate or,e tumad. to these provinces with dommstrated greater impimetatie capaity. Katimated ProieUS Coastes (in m mild. of Ja 196) A. Infrtructure 98.3 42.0 140.3 S. comnity Facilitie 12.8 5.2 18.0 C. Project Design and Supervision 9.4 1.7 11.1 D. Fquipmsnt and Vehilel 4.8 18.8 23.6 E. Technical Assistanc- and Training 6.7 3.4 10.1 Total Base Cost 132.0 7i.1 203.1 Physical Contingencies 13.2 7.1 20.3 Price Contingencies 10.8 3.8 16.6 Total Contingencies 24.0 12.9 36.9 TOTAL PROJECT COST 16084.0 240.0 Finramina Plan Bank 36.0 84.0 120.0 Provinces 32.0 0.0 32.0 Municipalities 88.0 0.0 88.0 Estimatd Disbursemsents (Bank Fiscal Year) 1988 1989 1990 1991 1992 1993 1994 1995 Annual 5.0- 9.7 18.3 25.8 23.0 19.0 16.1b 3.ib cumulative 5.0 14.7 33.0 58.8 81.8 100.8 116.9 120.0 Economic Rate of Returns 381c * Including initial deposit of US$ 5 million Into the Special Account, and retroactive financing. b Rate of disbursement reduced to recover advance =ade for Special Account. c Weighted average calculated for components comprising 581 of total project costs. I. THE SECTOR Recent Urbanization Trends 1.01 Argentint is one of the most urbanized countries in Latin America, with the unique characteristic that urbanization vas well advanced before World War I. In 1980, 832 of the total population lived in urban arsas, defined as settlements of 2000 or more inhabitants. Although it has not experienced the rapid urbanization in recent decades of some other countries of the region, Argentins Is still experiencing a significant Increase in the growth of its urban population. While the total population grew at a rate of 1.82 per year from 1970 to 1980 (23.4 to 27.9 million), the rural population decreased (losing almost 500 thousand from 5.1 million in 1970), and the urban population grew at 2.4% per year, to 23 million Inhabitants. Argentina's census bureau (INDEC) *stimates that the urban population will reach 28 million by 1990. 1.02 Argentina also is one of the Latin American countries where the urban population is most heavily concentrated in its capital: in 1970, the Buenos Aires Metropolitan Area (AMBA), with 8.5 million inhabitants, contsined 362 of the country's total population and 462 of its urban population. The two largest urban agglomerations outside the AMM, with 1970 populations in the 0.5 to 1.0 million range (Cordoba and Rosario), represented 72 of the total population in 1970. Over the 1970 to 1980 period, however, there has been a trend towards decentralization of growth in urban population: the AMBA population decreased from 36% to 352 of total population; urban agglomerations in the 0.5 to 1.0 million inhabitants range increased in number from 2 to 5 (including Mendoza, La Plata, and San Miguel de Tucuman), and accounted for 122 of the total population; and cities of less than 100 thousand population proved to be dynamic, increasing their share of total population from 23Z in 1970 to 262 in 1980. In terms of absolute numerical growth of the country's urban population over the 1970180 period, 2.2 million inhabitants were added to the capital and five largest urban agglomerations, and cities of less than 500 thousand grew by 2.7 million inhabitants. 1.03 These demographic trends reflect a decentralization of economic activity. While the Gross Regional Product (GRP) of the country as a whole grew by 28.5Z over 1970180, it only grew by 162 and 182 in the Federal Capital and the Province of Buenos Aires, respectively. This contrasts with growth rates over the same period of 1232 for Neuquen and 432 for Cordoba. This decentralization of economic activities is reflected in the more recent data on manufacturing employment. Over 1975/85, the country's manufacturing employment decreased by 102, representing a loss of 152 thousand jobs. In the AMBA alone, however, 166.6 thousand manufacturing jobs were lost, while there was a net gain of 14.6 thousand jobs in the rest of the country. In many cases, increases in manufacturing employment have been significant, particularly in the "frontier' areas, such as Neuquen and La Pampa, where manufacturing employment increased over 1975185 by 87.52 and 46.52, respectively. Government Structurl for Provisino of Public 8rvye. 1.04 Tho Argentine State 1t organised to three political-admindstrative and territorial jurisdictions: the contral, prowlacial and uanicipal lev.l.. Within this federal orgaisatiom, all govwrrAnntal power formally *mante from the Province. Province. delegate l mtod powers to the Cbetral Oo-rommt aNW retain for thoselves a11 other authority. Province are autonomus entitie governed by their own Constitutions. Municipalities are organised as part of each Prowince's system of government. Although the Natioal Constitution establishes that uniclpalitte should be indepedet from other loev of Goaverment, It also requires each Prowc to establish a Municipal Cod., In which spocific responsibilitie antd authoritis are delegated from Prowlace to Municipality. 1.05 The Central Government ti empowered to meet the main general needs of the Argentina society and to provide for the comon Interest of the different territorial subdivisions. There are also "concurrent powers entrusted by the Constitution to both the Central Governmnt and the Provinces, and even to Municipalities by provincial delegation. Because of the diversity of Provincial Constitutions, and a historically strong Interpretation of the Central Government's authority in practice, there ia provincial diversity, and significant areas of ambiguity, regarding service delivery authority and responsibility. In general, however, responsibilities for service provialon may be Illustrated as followst exclusively Central Government -- defense, foreign affairs, inter-Province transportation and trade regulation, mail and telecommunications; Central and Provincial Governments -- secondary and higher education, preventive health, justice, security, economic development, major passenger and cargo terminals, housing, electric and gas energy; Provincial and Municipal Governments -- elementary education, primary health care, water and sewerage, regional and local roads, fire control; and predominantly Municipal Governments -- solid waste collection and disposal, local streets and dralnage, parks, markets, cemeteries and land use planning and control. Public Sector Financln ond Interovernmental Fiscal Relations 1.06 During the 1960s and 1970s, the fiscal significance of Argentina's 23 provinces and territories grew substantially. Real expnditures Increased rapidly, both in per cggita ters and relative to total economic activity. Overall, lor capita provincll/municipal xpenditures grew twice aS fst as central government outlays. In terms of GDP, provincial/municipal expenditures nearly doubled from 5.42 in 1960 to an average of 10.2Z In 1977180. In the latter period, provincial expenditures exceeded central government expenditures by wide rgins in the ares of health and general admnistrative overhead. For the area broadly classified as economic developmnt, the central government's expnditures have been decreasng (from 6.12 of GDP In 1960 to 4.2Z In the latter 1970.) relative to those of provincial and municipal governments (1.4S of GDP in 1960 to 2.42 in the latter 1970a). This trend has continued into the 1980., with provincial/municipal expenditures representing 11 0% of GDP for the 1984185 period, as compared to the Central Government's average of 8.82 of GDP for the same period (excluding public enterprise. end social security, which together averaged 32.22 of GDP for the period). -3- 1.07 As provided in the Ntional Constitution, the Central Goverument has the following sources of revenues: (a) resources allocated to it exclusively and on a pen ennt basi (taxes on foreign trade and profits from the postal service)I (b) resoures allocated permanently in conjunction with the Prowines (dousetic taxes); ad (c) resources allocated temporarily and on a shared basis with the Provinces (direct taxs in cases of national emergency). Categories (b) and (c) bave contributed approxinately 802 of total Central GoverrAsnt tax revenues, and they bhav formed the base for Central/Provincial Goverrmants' revenue sharing system. It *bould be noted that the formal central/provincial revenue sharing systm ws suspended In 1985, an new legislation currently is being forulated. For 1985/86 and util the new systm is formalised, the shared revenue base for 1984 has been used, with adjustments for inflation, for each province. Thus, the former revenue sharing alloc-tion criteria described below (with details In Annex 1) are still b"alcally in place. 1.08 Under this revenue *having system, 31 of the total collected went to a Regional Development Fund (a special fund to promote regional integration), with the rest divided equally between Provinces and the Central Government, Including 1.81 of the total for Buenos Aires Municipality and 0.21 for the Territories. The formula for distribution among the 22 Provinces was as follows: (a) 65Z in direct proportion to population; (b) 251 In direct proportion to the development gap' (based on indicators of housing quality and number of automobiles per 1,000 Inhabitants) between each Province and the most developed area; and (c) 101 among Provinces that have population densities below the national average, In Inverse proportion to the difference between the density of each eligible Province and the average. The impact of this formula has been strongly rodistributive. In 1980, for oxample, the Province of Buenos Airos, which was allocated only 28.31 of local total rovenue shared usiug this formula, would have received 43.51 of this total if population were used as the criterion. The population criterion Itself, of course, is generally redistributive, unless there is a strong, positive corrolation between tax roceipts and population, which is rarely the case. 1.09 The preceeding paragraph briefly describes the legally determined system of revenue sharing botween Central and Provincial Governments (conarticinacion federal), In place formally to 1984, and informally since then. In recent years, particularly as domestic taxes have fallen with the national recession, coparticipacion transfers have decreased in real terms. To partially take up the slacks in budgetary resources at the Provincial level, the Central Government ha Increased its support to Provinces through ad hoc, discretionary transfers (acuerdos financieros). These discretionary mechanisms have resulted, however, in total transfers to Provinces that vary widely from year to year. For example, over the 1980- 84 period, total revenue sharing and discretionary grants transferred from the Federal Government ranged from US$3.5 billion in 1980 to US$1.9 billion in 1982, and finished the period at US$2.7 billion. As the provinces share the amounts transferred from the Federal goverament with their municipalities, this temporal instability at the highest lovel is transferred down to the local level, making it very difficult or impossible to plan, especially for capital expenditures. The impact of this temporal instabillty at the municipal level depends on the degree of provincial dependence on revenues derived from Federal sourcea (including royalties for extraction of natural resources). This dependence varies greatly. In 1985, for example, only 37.51 of the total revenue of the Province of luenos Aires was from Federal sources, versus 56.21 for Santa Fe, 61.22 for Cordoba, 77.62 for La Pampa and 82.32 for Neuquen. 1.10 Intergoverrmental fiscal relationships between provinces and municipalities are similar to those between central and provincial goverements, although each Provincial Constitution provides for specific arrangements. Fiscal transfers from provinces to municipalities may be automatic (revenue sharing, or copgrticl2acion) or discretionary. In principle, revenue sharing should provide resources in a sufficient amount so that, combined with a reasonable municipal fiscal effort, municipalitles would at least continue to maintain their current level of services, and the need for discretionary transfers would be minimal. Further discussion of the inter-governmental transfer system is presented in Annex 1. 1.11 In general, provincilX/municipal revenue sharing allocation criteria are predominately redistributional. Even in the two proposed provinces that stress devolution (La Pampa and Neuquen), over a quarter of the allocation criteria are redistributional. In the Province of Cordoba, for example, 10% of the provincial revenues from property and sales taxos, as well as of Federal covarticipacion are transferred to municipalities. Under the new (1987) Provincial constitution, 1% of this total amount will be set aside for a Municipal Development Fund. The remainder is then divided into three parts: municipalities receive 95Z; 32 goes to Comisiones Vecinalesl ; and the reserve fund receives 22. The criteria used for the allocation of the 952 share to municipalities under the existing law2 are highly redistributional: 76.8% is allocated among the municipalities according to redistributional criteria (population, 40.32; equal shares 18.2Z; inverse of population growth, 3.32, and personnel costs, 15Z).3 Thus, although 23.2% of the total coparticipacion transfers to municipalities provide some incentive for municipal tax effort, the main thrust is compensatory or redistributive, tending to allocate relatively higher per capita transfers to municipalities with smaller populations, slower growth and weaker fiscal bases. 1 12 The primary municipal own-source revenues are: user charges for public services (frequently tied to the property tax) such as street lighting, solid waste collection and disposal, storm drainage, and street maintenance; and betterment levies for public works that directly enhance property values. The percentage of total municipal receipts originating from fiscal transfers from provinces increased from 382 of total revenues in 1980 to 511 in 1983, while their real value contracted and fluctuated It Villages with insufficient population to become municipalities. d cThe regulatory law establishing the new criteria for allocation after the constitutional change has not yet been enacted, but the proposed law is practically the same as the existing one. 31 For more details, see Annex 1, especially Table 16. -5-~~-- widely from yecr to year. As in the Central/Provincial rclationship, the Provinclal/Mlnicipal trasnfor system ha tended to respond to general economic conditions. As the formal revenue sharing transfers have decreased In the 1980s, Provinces have tried to compensate at least partially for this with increases in discretionary trnsfers to municipalities. In the Province of Cordoba, covartielDlones repreoeted 952 of total transfors to munlciplities in 1980; in 1983 theso decroesed to 502 and in 1984, coparticivaciones were 702 of total transfors. Thus, total provincial tranfore to municipalities have varied greatly, not only in amount, but also by type (legally mandated vs. discretionary). The unpredictability of provincial transfer receipts, combined with unfavorable tera and very limited access to borrowings, has made development of longer-term municipal capitol improvements programs practically impossible. Sector Development Issues and Obiectives 1.13 The Sector's development issues and objectives are closely linked to national issues and objectives, primarily those of enhanced resource mobilization and increased public sector efficiency. A key national objective is to mobilize resources for needed public investments and recurrent costs in a manner that will reduce deficit financing and inflationary pressures* An effective way to meet this objective is to tranfer, to the maximun extent possible, the full financing burden for public investments and recurrent costs to their direct beneficiaries. The central and provincial governments have limited potential to recover costs directly from beneficiaries. At the municipal government level, however, there is substantial potential, frequently underutilized, for direct cost recovery from beneficiaries. Furthermore, as Government's decentralization policy proceeds, this potential should increase. Instruments for improved municipal direct cost recovery include betterment levies, connection fees, user charges, concessions, rents, etc* To the extent that these direct cost recovery mechanisms may be improved and expanded, municipal governments may contribute substantially to national objectives of resource mobilization in a non-deficit and non-inflationary manner. 1.14 The resource mobilization objective is closely linked to the objective of improved public sector management. To effectively mobilize resources, municipal governments must enhance their capacity to plan, program and budget for capital improvements programs, as well as for operation and maintenance of existing and proposed facilities. Effective decentralization and improvement of public sector management at the sub- national level greatly depend on better financial and service delivery responsibility relationships between Provinces and Municipalities. The Argentinas: Economic Memorandum (1985, p. 95) identifies at least three key issues for successful implementation of these improvementst (a) how to achieve greater predictability of, and control over, the volume of annual transfers to the provinces and municipalities; (b) how to achieve better public accountability for the resources transferred; and (c) how to encourage provincial and local government reforms in planning, budgeting, financial programming, etc. Sector Lending Strategy and Rationale for Bank Involvement 1.15 Through increasing the resource base and service delivery efficiency of municipal governments, the Bank would support two of the -6- broad objectives for which the Gbovernent he. solicited Bank supports Increased public aector efficiency; and resource sobilisation. The lank's support would focus on improving the financial managemont relationship between Provinces and Hunicipalities, particularly In promoting effective municipal capital Improvements progr _ that offer adequate cost recovery, priuarily from the program's direct benefelries. Through promoting fiscel disciplin and an appropriat, balance betwee expenditure responsibility and revenue authority by level of governmnt, the Bank would support resource mobilization In a financially responsible smnner. II. TIMROtC Project Oriain 2.01 The proposed operation would be the first of Its kind in Argent m. It builds on the operational experience gained In two proj-ets being iplementod by the Brazilian States of Parnc and Santa Catarina (Loon Nos. 2343-5R and 2623-U). The project would benefit the Provinels of Buenos ALres, Cordoba, La Pma_, Neuquen and Santa Fe, and If the experlence proves successful, a sequence of further operations would be developed to benefit other Provinces. Pro10ct Obi ectives 2.02 The main purpose of the proposed project Is to contribute to wor effective public sector management at the Provincial and Municipal levels of goverrment through Improved finaneing mechanisms for justified municipal investments. The specific objectives of the project would be to: (a) mobilisz external and internal resources in a non-deficit and non- lnflationary way to finance municipal lnvestmnts with positive economic and social impact.; (b) strengthen municipalities' capacity to plan, finance and execute cost-effectiv* capital lmprovements progrms; and (c) promote structured, periodic consultations between municipal and provincial authorities for the formulation and ovaluation of Investment plns. Proiect Components 2.03 The project would finance the following component categories, as described in greater detall in Annex 2s (a) Physical investments (representing at least 902 of total project costs), including construction and rehabilitation of public infrastructure (e*ge, water, sewerag, street paving, storm drainage) and community facilities (e.g., bus terminals, health posts), and purchase of vehicles and equipment (e.g., road maintenance, garbage collection); and (b) Technical assistanco and training (generally representing up to 5S of total project costs in each province, or up to 10 in exceptional cases with justifications satisfactory to the Bank), including financial management, accounting procedurs, cadastres, maintenance proeedures, computers, supervision vehicles, etc. Criteria for Implementation of Project Objectives 2.04 In order to effectively implement the project's objectives in five distinct Provinces, each with unique demographlc, economic, physical and political characteristics, it is essential that the basic criteria necessary for effective implementation of project objectives be clear, -8- easily determined and fully accepted by all institutions Involved In the project's implementation. These basic criteria, covering (a) eligibility for (i) Provincial (ii) Municipal and (iII) investment component participation in the project and (b) the minimum conditions for capitalisation of a Municipal Development Fund In each participatiAg Province (providing for project replicability) must be uniformly defined end applied in all participating Provinces, as described below. 2.05 Regarding provincial eligibility, a fundamental condition for achieving the resource mobilization objective is that each participating Province has a revenue sharing (coParticipacion) system that: (a) has an acceptable balance of fiscal devolution and rodistributive criteria for allocation of the shared revenues; and (b) provides shared revenues in a sufficient aount so that, combined with a reasonable municipal fiscal effort, municipalities may at least continue to maintain their current level of servico. The revenue sharing systems of the five proposed Provinces are described in Annex 1, Part C, and the above conditions have been met in all of the proposed Provinces. 2.06 Regarding munmicial eligibility criteria, only municipalities in which investment decisions for the kinds of infrastructure to be financed under the project are substantially independent of investment decisions of other municipalities would be potentially eligible to participate in the project. This criterion is based on the following considerations: (a) the project seeks to strengthen municipal institutional capacity to more effectively provide municipal services; (b) it also seeks to strengthen a significant number of municipalities through financing numerous, small- scale municipal investments; and (c) to include metropolitan areas and other urban agglomerations, with complex inter-municipal infrastructure systems, would jeopardize the rapid resource mobilization objective through more complicated technical issues and institutional arrangements. Eligibility of municipalities according to urbanization characteristics is determined in Annex 1, para. 6, and Attachment 1. 2.07 Within the above criterion, all municipalities would be eligible to participate in the technical assistance and training component of the project. In order to receive credit to finance physical investments, however, municipalities must meet creditworthiness criteria. To be eligible, a municipality must demonstrate, initially and during the entire project execution period, that: (a) its current revenues (including municipal own-source revenues and legally regulated revenue sharing, cokarticipaciones, but excluding discretionary grants) are greaterl than its current expenditures (including payments of interest and principal); and (b) its outstanding and proposed debt obligations are within reasonable limits, defined by the following criteria: (i) total outstanding and proposed debt not to exceed 60X of total municipal revenues (excluding credit) of the previous year, duly adjusted for inflation; and (ii) total debt service on total outstanding and proposed loans not to exceed 151 of total budgeted revenues for the year of the proposed project. 1/ For details on the application of this criterion, see Annex 2, para. 3. -9- 2.08 Regarding comoonent eliuibility criteria, * fundamental condition for achieving the objectives of the project is that e*ih annual investment plan presented by each municipality has at least 652 of total proposed physical investments allocated to components for which 1002 of total investment costs will be fully recovered from payment to the municipality by direct beneficiaries (betteor nt levies, connection fees, ueer charges, sales, concessions, rents, *tc.). This criterion, together with the municipal creditworthy criteria (para. 2.07) would provide adequate assurances thats (a) incremental resources mobilized through the project would be used An a non-deficit and non-inflationary way; and (b) municipalities that psrticipate In the project would be financially strengthened in order to provide other s*rvices not financed under the project. 2.09 The municipal creditworthy and component eligibility criteria noted above would provide the conditions necessary for project replicability within each participating municipality. The project also envisages replicability across municipalities. This would be achieved through creation of a Municipal Development Fund in each participating Province as a condition of disbursement (para. 3.03(a)), or more generally, as a condition of loan effectiveness (para. 3.02(a)). All of the proposed Provinces have agreed that 'harder' onlending terms to municipalities are necessary to capitalize the Municipal Development Funds. However, legitimate doubts exist regarding the 'ideal' level of Fund capitalization in each Province, because there are doubts about sustainable levels of municipal borrowings from the Fund, particularly in light of possible future, alternative sources of borrowing that are not defined now. To ensure a minimum level of project replicability, resouces capitalized in these Funds would be used exclusively for further onlending to municipalities at positive real interest rates. Furthermore, it has been agreed with all Provinces that each would be autonomous to establish its onlending policies to municipalities within the following minimum measures for-Municipal Development Fund capitalization (para. 3.01(b)(i)): (a) while project loans from the Central Government to Provinces would be repaid over 15 years, including three years of grace (para. 2.22), Provincial onlending to Municipalities would have up to a maximum of 10 years for loan repayment, including up to a maximum of one year of grace; and (b) in onlending to Municipalities, Provinces would include an interest rate spread of at least 1.5 percentage points over the rate for Provincial repayment of loans to the Central Government. 2.10 These arrangements would ensure capitalization of Municipal Development Funds in each participating Province of at least 33S, as measured by the present value of net funds flowing to the Fund over 20 years, and expressed as a percentage of the initial loan withdrawn over three years. From an alternative perspective, these arrangements would capitalize each Fund at least to the level of 131, as measured by the - 10 - annual balanc estimated to flow to the Fund in year 18 and beyond, and expressed as a percentage of the average annual loans financed under the project for the Initial thre year implementation period. From the perspectivo of project beneficiaries, who mst finance the capitalization of the Municipal Development Funds, the 1.S percentage point spread would represent, on average, an increase of approximataly 62 In annual beneficiary repaymens to municiplities with a nie year amortization period. Further analysis of Municipal Development Fund capitallsation Is presented in Annex 4. 2.11 As the political jurisdiction closest to popular pressures for improved public infrastructure and services, municlpalities ft-quently serve as interediaries" for services that tbey are not mandated to provide. This is the case, for example, when a municipality uses the betterment levy to recover the investent costs of extending so ggas networks to residential areas; after construction ls eompleted, however, the municipality transfers these Infrastructure ass*ts to the national ga company nd has no authority over the gas company's operation. If the proposed project financed such Investments, it would be supporting the national gas company rather than strengthening the municipality. Thus, In order to meet the project's objective to strengthen municipalities' capacity to plan, finance and execute appropriate investment programa, the project would finance only those physical investments that are exclusively municipal In nature. The criteria for determining if physical Investments are exclusively municipal in nature are positive responses to all of the following questions: (a) Will the investments be procured under the municipality's authority? (b) After the Investment is completed, will the municipality be responsible for its operation and maintenance? and (c) After the investment Is completed, will the municipality have authority over service tariffs or fees? 2.12 In some provincs, service provider cooperatives are widespread, and often there are questions regarding final reponsibility, municipal or cooperative, for service provision. In these cases, investments ould be considered eligible for financing under the project when: (a) all of the three criteria of the proceeding paragraph are satisfied; and (b) municipal authority is legally delegated, with limitations of delegation authority, to cooperatives In which participation is restricted to persons or firms located within the jurisdiction of the munlcipality. 2.13 In order to (a) promote broad institutional benefits resulting from implementing numerous, small-scale investments In many municipalities and (b) be consistent with simplified methodologies for economic (Ann 5) and technical (Annex 6) appraisals, *ligible civil works subprojects to be financed In each municipality would be limited to US$ 500,000 equivalent. For application of this criterion, a civil works subprojoet Is defined as having all the elements necessary to be fully functional, either on its own, or in conjunction with other works that already exist. - 11 - Proiect Coats and Allocation by Proaince 2.14 The total project coat is estimted at US$ 240.0 million equivallnt, Including bae coats and contingencies, but excluding lead, dutles and taxes, which are lnlilgible for lank financing. The proposed lank loan of US$ 120.0 millieo equivalent would finance 50S of tbese totel costs. A *u Ary of project costs is presented in Table 1, and details are prsented In Annex 7, Tables 2-4. Additional details, with disaggregations at the Provincial leel, are available In the Project Fi1e. Table 1: ESTIMATED PROJECT COSTS (in USS millions of January 1988) F. .C. s Z of Component Local Forcian Total 2 of Total lDs* Costs A. Infrastructure 98.3 42.0 140.3 30.0 69.1 B. Comunity Facilities 12.8 5.2 18.0 30.0 8.8 C. Proj. Design & Sup. 9.4 1.7 11.1 15.0 5.5 D. Equip.& Vehicles 4.8 18.8 23.6 80.0 11.6 E. TA & Training 6.7 3.4 10.1 35.0 5.0 Total Base Cost 132.0 71.1 203.1 35.0 100.0 Physical Contin. 13.2 7.1 20.3 35.0 10.0 Price Contingencies 1 _ 5.8 16.6 35.0 8.2 Total Contingencies 24.0 12.9 36.9 35.0 18.2 TOTAL PROJECT COST 156.0 84.0 240.0 35.0 118.2 2.15 Project cost estimates have been made based on actual coats of similar Items recently procurod in the proposed Provinces. General and unitary specifications nd dimensions of typical subprojects have been reviewed and found acceptable to the Bank, and are presented in Annex 7, Table 1. Each Province's Project Operations Manual would include provisions, satisfactory to the Bank, for review ex-ante, as well as supervision of *xecution, by competent provincial authozities (Ministry of Public Works) for each sub-project proposed by municipalities (see Annexes 3 and 6). 2.16 Costs are expressed in US dollars because of the unpredictability of national inflation and devaluation rates during the implementation period. Blse costs were determined by updating dollar equivalent costs of Juno 1987 to those of January 1988, applying the expected yearly rate of inflation of 31 for 1987. Annual rates of inflation estimated for international price trends for manufactured goods and civil works were used to estimate the project's price contingoncies: 1988-90, 12; 1991 and thereafter, 3.52. Physical contingencies averaging 102, and price contingencies averaging 82, have been estimatd. -12- 2.17 For the total project, foreign exchange costs are estimated at 35, with approximately 302 for infrastructure and community facilities, 15S for project engineering and supervision, 352 for technical assistance and 801 for equipment and vehicles. 2.18 Disensioning of project and loan sizes are based primarily on (a) the debt limitations and actual investment levels over the past five years of creditworthy, and thus financially eligible (par&. 2.07), Municipalities of the five Provinces, (b) Municipal institutional capocity to expand their historical levels of investment and Provincial institutional capacity to support xpanded municipal investments and (c) the investment programs presented by each Province based on (a) and (b) above, as well as estimates of the actual municipal demand for credit over a project execution period of three years (for additional discussion regarding dimensioning of the project, see Annex 4). Based on the above, the total Bank loan would have an initial allocation by Province as follows (in US$ millions equivalent): Buenos Aires 55.0 Cordoba 21.0 La Pampa 3.5 Neuquen 9.5 Santa Fe 31.0 TOTAL BANK LOAN 120.0 2.19 The above amounts are those agreed with each Province as loans that may be 'reasonably' estimated to be absorbed over a three-year implementation period. However, as this would be the first municipal development project for Argentina, it would be prudent to estimate actual implementation over a longer period to accomodate the frequently observed delays in start-up, possible institutional discontinuities, etc. Thus, the formal implementation schedule would be based on the average disbursement profile for lines of credit in the region (7.5 years). 2.20 In addition to uncertainties regarding the total loan absorptive capacity over time, there are uncertainties about the relative municipal loan absorptive capacity between Provinces. In order to spread the risks of these uncertainties, as well as to reward those Provinces with greater project implementation capacity, the following arrangements have been agreed with Central and Provincial Government authorities. The initial allocation (para. 2.18) would be assured for the first three years of the project's implementation. However, the legal documents would establish that each Province could draw down 'up to' the stipulated amounts during the first three years of project implementation. At the end of the third year, if not all of the total loan has been committed (ie., a properly procured contract has been entered into for specific goods, works or services to be financed under the loan), then a new allocation of the remaining loan would be determined according to provincial performance during the first three years of project implementation (para. 3.01(a)(i)). The formule for this nw allocation would be as follow. -- The uncomitted loan amount, If any, remaining at the end of year three (UL3) is the total, accumulated, loan comittments at the end of year three (TALCS) subtracted from the total Initial loan amount (TIL): UL3 W TIL - TALC3. Each Province that elects to stay In the project would have the right to a part of UL3 during year four of projeet impl_mentation. The loan allocation for year four for Province (LA41) would be based on that Province's participation in total, accumulated loan conmitment during the first three years of project Implementation (LC3i)t LA4i - (LC3i1TALC3)*UL3. -- This procedure would be repeated each year until the proceeds of the Bank loan are fully withdravn. For example, at the end of year four, Province i's allocation for the next year (LASi) would be: LASL - (LC4i/TALC4)*UL4, where LC4i is the mount of loan commitments made to Province i and/or to its Municipalities during the four years of project execution, TALC4 is the total amount of loan commitments in four years, and UL4 is the uncommitted loan amount at the end of year four. Finsncint Plan 2.21 Because of the project's high social and institutional impacts (par-s. 2.44-2.45, and Annex 5), Bank financing of 502 of total eligible project costs is proposed. With all counterpart funding provided by Municipalities and/or Provinces, the project's financing plan would be the following: Amount Source USS Millions S Bank 120 50 Provinces 32 13 Hunicipalities 88 37 TOTAL 240 100 Except for very municipal-specific cases (see Annex 2), provinces would provide all counterpart funds for technical assistance and training. Because of the different intergovernmental distribution of service delivery responsibilities and revenue sharing policies in each Province (see Annex 1), different Provincial/Municipal 'mixes' of counterpart funds for municipal investments are justified. Thus, counterpart funding for municipal investments vould be as follow:s 1OOZ of counterpart from Municipal funds in the Provinces of Buenos Aires, La Pampa, Neuquen and Santa Fe; and 1001 of counterpart from Provincial funds in Cordoba. All expenditures ineligible for Bank financing (e.s., land and taxes) would be totally financed by Municipalities. 2.22 All proceeds of the Bank loan would be onlent from the Central to Provincial Governments with which the Bank has a Project Agreement (para. 3.01(a)(iii)) under the same term and conditions as Bank lending to the 14~~~~~~~~~~~~~~~~~~~~ - 14 - Argentin Republic (except as noted in par&. 2.23), to be established In Subsidiary Loan Agreements between the Central Government and *ach Province. A Special Account (para. 2.30), in US dollars and with an lnitial deposit of approximately US$ 5 million, would be established in a financial institution, satisfactory to the Bank, dsignated as Financial lntermediary (par- 3.O1(a)(i1)), under a Financial Agency Agreement with the Central Governmet which would be a condition of effectiveness (para. 3 02(b)). The Financlal Intemuediary would receive directly from the authorized representative, of the Provincial Governments certified Statements of Expenditures (80Ss), which would support the Financial Intermediary's requests to the bank for replenilabent of the Special Account. 2.23 The Central Government would assume the cross-currency exchange risk (par&. 3.01(a)(iv)) but would transfer to the Provinces the US dollar exchange risk (par&. 3.01(b)(ii)). Thus, Provincial debt servicing to the Central Governmnt would be denominated in terms of the official (coimercial) exchange rate of the US dollar. In order that the dollar exchange risk, over which the Provinces have no control, be made reasonably predictable for Provincial financial planning purposes, the periodic, dollar-indexed provincial debt service adjustments would be restricted to a 'band' limiting dollar-indexed debt service adjustments to a plus or minus IOZ deviation from the official inflation rate (as measured by INDEC's General Wholosale and Consumer Price Indices, weighted equally). If and when the dollar-indexed, provincial debt service falls outside this band, then the difference ould be credited to (or capitalized in) the provinces' outstanding loans. In addition, each province would pay a charge, for services of the financial Intermediary and of the National Liaison Unit (par&. 2.26), of up to 12 on loan amounts disbursed (3.01(b)(ii)). 2.24 Onlending of tho proceeds of the Bank loan from Provinces to Municipalities would be on terms and conditions established in subloan agreements between each Province, Its Provincial lank and Municipalities. In order to capitalize a Municipal Development Fund in each Province to a minimum level necessary for project replication, these terms would be 'hardened at least as much as indicated in pars. 2.09. 2.25 Provincial Banks would serve as the provinces' financial agents for administration of municipal subloans. Also, proceeds of the Municipal Development Funds would be deposited In the Provincial Banks, under the administrative authority of the Provincial Project Execution Units. Municipal repayment of loans would be guaranteed through the Province'. retention of the Municipality's debt servicing obligations from Provincial revenue sharing. Indexation of debt service payment by the Municipalities to the Provinces would be according to the official US dollar exchange rate (with the same wbands limitations noted in pars. 2.23) in all Provinces except for Buenos Aires, which would follow official INDEC price Indices. The precise method of indexation of municipal subloan debt service payment is not a fundamental condition of project implementation, so long as a positive, real interest rate compatible with the opportunity cost of capital in Argentina is achieved. An indexation either according to the US dollar or to an adequate indicator of internal inflation (such as INDEC price indices) Is acceptable, and provincial autonomy to choose en acceptable indexation lnstrument is compatible with project objectives. 15~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - 15 - 1luDemntation Arraneemente 2.26 Overall supervision responsibility for the project wvuld rest with the Borrower's Ministry of Economy, but technical supervision would be delegatod to the Secretariat of Rousing and Environmental Managemnt (SVOA), Ministry of Public Health and Social Affairs (par.. 3.01(a)(v)). The role envisagd for SvoA is uodest, *seentially limited to monitoring and reporting on project Implementation progress, and it would not require more than two to three professionals. A key function of SVOA, through the project's National Liaison Unit, would be to receive the *smi-annual performance and annual financial audits from each Province (pres. 2.38- 2.39), to consolit-te these into a project-wide evaluation on a semi-annual basis, and to submit at least semi-annual progress reports to the Bank for supervision purposes. The proposed monitoring and reporting experience of SVOA would Improve Its capacity to transfer the lssons learned from the proposed initial operation to further operations benefitting other provinces. 2.27 Each Province would form a Project Execution Unit (PEU) that would be primarily responsible for implementation, supervision and evaluation of the Provincial subproject (for details, see Annex 3). The PEU would not be a large entity, duplicating the functions and tasks of oxisting agencies. Rather, it would function under the authority of the Governor to ensure that, in addition to performing directly all planning, programing, supervision and reporting duties specified In the Project Agreement, all provincial ministries and agencies necessary for successful project Implementation are properly coordinated (par&. 3.O1(b)(iii)). The Bank has reviewed each Province's proposal for its PEU and found each to be acceptable. However, in addition to the agreed-to general design, PEM staffing satisfactory to tho Bank would be a condition of disbursement in each Province (par&. 3.03(a)). Furthermore, the Bank's supervislon would include, with SVOA's assistance, evaluation of the onnual performance of each Province's PEU, including structural and staffing aspects (pare. 2.38). If significant deficiencies are identified, then technical assistance for the PRU, directed at structural and/or staff strengthening, would be an obligatory part of the subsequent year's investment program to be agreed with the Bank (par&. 3.01(b)(viii)). 2.28 Municipal authorities would be Invited to propose annual, or multi-year, investment plans to the PEU for Inclusion in the province's program proposal for the subsequent year. This would take place through a structured process of consultation on a rolling, annual bais. Annual Investment Plans, as consolidated and proposed by the province (including provincial technical assistance and training, as well as municipal Investments), would be reviewed and approved by the Bank (pare. 3 .01(b)(iv)), and only subprojects included in these Plans would be eligible for financing under the project (pare. 3.03(c)). 2.29 Detailed ip lementation arrangements would be described In each Province's Project Operations Manual, which each Province would agree not to amend without the Bank's prior consent (pare. 3.01(b)(v)). In addition to the project's eligibility criteria (paras. 2.06-2.13), tho Manual would include, inter alia, subloan terms and conditions, guidelines for procurment, draft tender documents, subloan disbursement procedures, _ 16 - guidelines and control procedures for compononts' unit costs, and model contracts betw"n Province, Provincial Bank and Municipality, as well a the text* of the Project Agr eme nt with the Bank and of the Subsidlary Loan Agrement with the Central Governmnt (further details In Annex 6). Procur m_nt 2.30 About 400 unicipalities in the five Provinceo *rc expected to participate In the projcet, resulting In a large number of sm*ll and scattered civil works, each with a maxium cost of US$ 500,000 equivalent (para. 2.13). Procurement for goods and works costing US$ 25,000 equivalent or less would be done through international or local shopping, Involving at least three price quotations. This is expected to account for approxImately US$ 30 million equivalent, or 131 of total proj3et costs. Procurement of civil works costing between US$ 25,000 and 500,000 equivalent would account for approximately US$ 139 million equivalont, or 581 of total project costs, and would be made through Local Competitive Bidding (LCB) procedures acceptablo to the Bank. Procurement of goods costing more than US$ 25,000 but less than USS 250,000 *quivalent would be through LCB procedures acceptable to the Bank, and would reprosent approximately USS 47 million, or 202 of total project costs. Procurement of all goods costing US$ 250,000 equivalent or more would be through International Compotitive Bidding (ICB) in agreement with the Bak's Guldelines for Procurement, representing about USS 2 million equivalent, or 12 of total project costs. Consultant services, estimated to cost US$ 10.2 million equivalent, or 42 of total project costs, would be procured in accordance with the Bank's guidelines. These arrangements are summarized In Table 2 (next page), and the Project Operational Manual in each Province would include the necessary provisions to ensure that procurement procedures are fully compatible with the Bank's Guidelines on Procurement (see Annex 6 and paras. 3.01(b)(vi) and 3.02(a)(iv)). 2.31 Force Account procedures would be used, on an excimptional basis, by municipalities to oxecute directly civil works, with total force account procurement of materials not to exceod 102 of the total cost of civil works. in each province on an annual basis. Moreover, the total cost of materials purchased to support a given civil work executed undor force account would not exceed US$ 150,000 equivalent, and would be purchased in *ccordance with the procedures outlined in pars 2.30 above. 2.32 Provincial PEU's would review, ex-ante, all procurement procedures, documents, bid evaluations and contract awards to ensure that the agreed procurement process is properly carried out. In addition, all ICB procedures, documents, bid evaluations and contract awards would be reviewed ex-ante by the Bank. Because LCB bid packages would be too small and numerous for effective, ex-ante review by the Bank, such review would be delegated, primarily, to each Province's Executing Unit. However, in order to ensure compliance with Bank procurement guidelines, each PM would submit, during the first year of project execution, six randomly selected LCB procurement procedures (three for civil works and three for goods) for ex-ante Bank review, Including bid documents, evaluations and contract wards. Furthermoro, the performance audit terms of reference (para. 2.38) in each province would provide for certification thrt the procurements presented to the Bank for ex-ante review are typical and representative. -17- If, after the first year of project inpiementation, the lank's supervision indates that the agreed procurement procedures are being followed eatisfact.rily, then the Bank's ox-ante revlew of W I bid procedures would be reduced to two (one for works, one for goods) per year per prow nc. In all caese, howaeer, the lank would review ILC, International end local shopping, and force account procurement proceduroe on a ex-yt basis by s* ling, and If it wore determined that procursenat we not made folloowi agreed procedures, tbhn no expenditure for such Ite_ would be financed out of the proceds of the loan, and the lank would canel the corresponding loan amount. Table 2: PUOCURDENXT NUS?OO Project Eloment Procurement Nthotd Total ICR ICI Int./Local Other N.A. Cost Shonlina ---------------e(US$ illion)--------------- Civil Works 139.2 17.0 11.8 168.0 (73.6) (9.1) (0.0) (82.7) Epulment. Materials and Vehicles -Hun. Infrast. 2.0 46.0 12.0 60.0 (2.0) (23.0) (6.0) (31.0) -Prov. PEUs 1.0 0.8 1.8 (0.5) (0.4) (0.9) Consultina Services -Prov. PEUs 9.6 9.6 (4.8) (4.8) -Auditing 0.6 0.6 (0.6) (0.6) TOTAL 2.0 186.2 29.8 10.2 11.8 240.0 (2.0) (97.1) (15.5) (5.4) (0.0) (120.0) Notes: Figures in parentheses are the respectiv, amounts financed by the Bank. The N.A. refers to legitimat- project costs for engineering design and supervision of civil works (estimated *a 72 of total works' cost), but the Bank would not disburse against this category. Disbursement 2.33 For withdrawal of the proceeds of the proposed Bank lo-n, the following categories and disbursement percentages are proposed: (a) for eligible civil works, 53.5Z disbursement against contractors' certified receipts (including as legitimate project costs municipal expenditures for engineering design and supervision of works, estimated as 71 of total contracted civil works - is - *xpendIturs bhowver, the withdrwal applications would not include direct expenditures for design end supervision, and neither would be eligible for Bank financang under consulting servio.a)l (b) for equipment, aterials and vehicles, 502 disbursement for local expenditures and 1002 of foreigp experAltures; (c) for consulting services, 502 disbursemnt for local expenditure and 1002 of foreig expenditure; and (d) for auditing sorvices (ih e provided by a private firm under a contract satisfactory to the Jank), 10O2 disbursament for local or foreign expeaditure. 2.34 For municipal expenditures in categories (a), (b) and (c) above, disbursemats would be authorised only after verification and certification, by the PEU according to procedures established In the Operations Manual, that ependitures resulted from a previously approved subloan agreement between the Mnicipalityr, the Province and the Provincial Bank, as the Province's financial agent. For Provincial expenditure In categories (b), (c) and (d), disbursements would be authorixed only after certification by the PEU that exponditures resulted from a technical assistance program (bnefitting the Province and/or Municipalities eligible according to par&. 2.06) acceptable to the Bank. 2.35 Because of the large number of contracts and relatively small payments, all disbursements for local expenditures and disbursements for foreign expenditures up to US$ 250,000 equivalent would be agaist SOEs. These SOEs would present expenditures by subproject with a breakdown of outlays for equipment, materials, consultants' services and civil works, either under contract or by force account. Documentation for these expenditures would not be submitted to the Bank but would be retained by the Provincial Executing Unit for periodic inspection by Bank staff. Disbursements for foreign ependitures of US$ 250,000 or more would be made againt standard documntation. Each Provincial Executing Unit would maintaln adequate records of muicipal project ependitures, and ach would prepare consolidated SOs for periodic transmittal to the Bank via the Financial Intermediary for the Special Account (par&. 2.22). 2.36 In order to encourage and reward provincial initlative for rapid and effective Institutional support for project Implementation, retroactive financing is recocm_nded, beginning February 1988, for those Provinces that (a) have a legally constituted Project Execution Unit, with a functioning staff acceptable to the Bank, (b) comply with all relevant procedures established In an approved Project Operations Manual, satisfactory to the Bank, and (c) comply with all other terms and conditions of the Loan and Project Agreemnts. Retroactive loan financing in any Province would be linited to 10 of that Province's Initial allocation of the total loan amount (pars. 2.18). 2.37 The estimated disbursement schedule follows the regional disbursement profile of 7.5 years (pars. 2.19), and this schedule is presented In Annex 7, Table 5. Assuming expenditures eligible for - 19 - retrosotive finaning beginnin In ebruary 1988 sa los effectiveness by late Jln 1988, than the project's ompletlon data mid be Deember 31, 1994, sa the proposed closin date would be June 30, 1995. It to expected, bouer r, that the proj et wod be executed nsot rapily the the regio al profile (Pares. 2.19-2.20), end a hborter exotimo periot would inrese the bef its of the project *ad Ladiost tht the Povimwe bhao potential to mobllise efficiently more resource for aditiomal lntrestructure lnvestmets. Accounts ad Audits 2.38 Separate Project Agreements midul be mad* with each Province establishing Its respossibilitles, including that of maintaining a separate Project Account. The Sp eial Account (pars. 2.22) and the Project Account In each Province and ia each participating Municipality miud be audited by external, Independent auditors acceptable to the Bank. for Munlelpalities and Provincial Executing Units, an annual audit of the financial statements of the Project Accounts, and including a separate opinion on withdrawal applications made on the basis of Statment of txpenditures, would be furnished to the Bank not later than four montbs aftor the end of each fiscal year. In addition, each province would provide to the Bank s*mL- annual performance audits within 60 days from the date they are lssuod. Thee performance audits would provide evaluations of the compliance by all participating provincs and municipalities with all legal agreements, including provisions of the Project Operations Manual, reched with the Bank, province, provincial intermediaris, and mnicipalities (par-. 3.01(b)(vii)). The National Liaison Unit within SVOA would consolidate all provincil financial and performance audits and provide an overall project progress report to the Bank on a semi-annual basis (para. 2.26). Audit of the Special Account mould be done according to standard country-wide procedures (para. 3.01(a)(vi)). 2.39 Special, *emi-annual, performance audits are essential to project design in order to make effective Bank supervision manageable In the fce of a large number of sub-projects spread over an enormous geographical area and within numerous political and administrative jurisdictions. BecaUse of the demndlng audit requirements of the project, and of the need to "ssure timely compliance with udit agreements and to minimise risks of Interruptions of disbursements through SO procedures, the Bank ould finance 1OOZ of audit xpenditure when thee are incurred by private firnm acceptable to the Bank (see pars. 2.33). If the flnancial and/or performance audits ldentify significant irregularities in any municipality, then the Project Operational Manual would provide that, as a condition of recelving project funding for municipal investments in the subsequent audit period, the province would provide a technical assistance progrem, acceptable to the Bank, focused on eliminating the observed lrregularities. And, if thes irregularitie persist for another audit period, that municipality may be declared ineligible for subsequent project financing (par&. 3.01(b)(viii)). Financial Impact 2.40 Municipal creditworthines and component eligibility criteria (paras. 2.07-2.08) vmld ensure that all municipalities that participate In K -~~~~~~~~~~~t - the project would be financially strengthened bccause of their participation. Dimensioning of the project has included cons iderations of historical levels of actual municipal investments (para. 2.18 and Annex 4) so that the ipact on -muicipalities receiving project loans would be to increae ,the rate and quality of service coverage in general and to xpand service delivery in poorer area in particular (soe Annex 5). For municipalities that will bo oligible initially only for technical saslitance, because they do not meet the creditworthinesi criteria, the project ould focus provincial asistance on Improving municipal financial performnce so that they may be eligible for physical investment loans in the future. 2.41 From the perspective of direct project beneficiarie, the project's financial impaet must be positive if two key asumptions are mets (a) benefitting families and firms are rational md accurate in estimating that proposed infrastructure benefits, as cpAptalized in their property values, will be greater than the costs thev rast pay (generally through betterment levies, sae para. 2.43, and Annex. !P) in order to receive the ser'7ice; and (b) beneficiaries agree to finance infrastructurk improvements on a voluntary bass. Because the experience with the financing of municipal infrastructure expansion in Argentina indicates that both of these assumptions are reasonable, the risk of negative project financial impact on direct beneficiaries is very low. 2.42 Evaluation of the project's impact on provincial finances is of broad concern only in the Province of Cordoba, which proposes to provide all counterpart funds for eligible project expenditures from Provincial sources (in the other provinces, all physical investment counterpart funding would be municipal, see para. 2.21). If the project is fully implemented in Cordoba over three years, then the annual demand for provincial counterpart funds would represent 62 of total Provincial expenditures for transfers (considering the 1986 executed provincial budget as representative), and 842 of discretionary grants specifically allocated to municipalities (see Annex 4, Table 2). In addition, Cordoba's new constitution (para. 1.11) provides for a special revenue sharing category: the 'Municipal Development Fund', which should not be confused with the project's Municipal Development Fund to be capitalized in each province through repayment of proceeds of the Bank loan (para. 2.09). One percent of all revenue sharing with municipalities will be allocated to this revenue sharing fund' and the 1988 provincial budget provides that all of this fund will be used as counterpart for the proposed project. In other provinces, the only provincial expenditures that will be required by the project are counterpart funds for the provincial technical assistance program (approximately 52 of total project expenditures) and the cost of operating the Project Execution Unit. Thus, Implementation of the project over three years would be affordable from the perspective of Provincial finances, and project implementation delays resulting from insufficient provincial counterpart funding would not be a significant risk if provincial political committment to the project continues (paras. 2.48- 2049). Economic Impact 2.43 The weighted average of the internal economic rate of return (IERR) estimates for all investment components evaluated (water supply, - 21 - sewerage, strest lighting and paving) In the five provinces is 38.52. The weighting mechanism used is the component's estimated cost a* a proportion of th. the project's total etimated cost. The components included in the economic evaluation represent 57.62 of this total estimated cost. The average Ihe range from 19.12 for *ewerag components to 84.42 for street lighting. Sensitivity analyses rovealed thats (a) increasing cost estimates by 202 lowered the average rate of roturn for all components in all provinces to 31.12; (b) lowering benefit estimates by 202 reduced the average IERR to 29.92; and (c) i!.creasing costs and reducing benefits by 20S resulted in an average IERR of 23.72. Only under this most pessimistic scenario would one component (sewerage) have an IERR approximately equal to the 122 rate of discount estimated as the opportunity cost of capital (Annex 5, par&. 8). Poverty Impact 2.44 The project would havo a significant social impact through: (a) financing an increase In 'adequate' service levels for water and sewerage (the lack of which is widely used as a poverty indicator In Argentina, see Annex 5, pras. 9-12); (b) providing increased affordability for these basic services, permitted by the project's relatively long-term credit financing, which would permit longer beneficiary repayment periods (Annex 5, Table 5); and (c) employment generation for un- and low-skilled labor. 2.45 In 1980, there were approximately 390 thousand households with linadequate" sanitary facilities in the eligible municipalities. This 'deficit" is equivalent to almost 1.5 times the number of households in Gran Rosario (the second largest metropolitan area in Argentina). Over half of the units with *inadequate' sanitary facilities in the five provinces and almost one-quarter of those in the country as a whole were located in the eligible municipalities In 1980. Bank financing for the project would permit significant increases in beneficiary repayment periods (up to two or three times the typical repayment periods of two to four years currently In practice). For example, increasing the repayment period from three to six years (for a given interest rate) would reduce the beneficiaries' monthly payment by nearly half. In general, beneficiary affordability is more sensitive to repayment period than interest rates a decrease in an annual interest rate from, for example, 122 to 8.52 would reduce monthly repayments by only 6% for a three year repayment period. Finally, the project would create approximately 58 thousand personlyears of construction work. Environmental Impact 2.46 The project would promote for the first time a systematic, inter- provincial, concern with the environmental impact of municipal infrastructure investments. Building on the relatively new environmental legislation introduced in the Provinces of Cordoba and Santa Fe, each Province's Project Operational Manual would include specific safeguards for environmental protection and enhancement. 2.47 The project would have a diffused and generally positive impact on the conservation and enhancement of the physical environment. Properly designed and constructed subprojects such as street paving, storm drainage - 22- ad landscaping produce significont reductions In erosion and, frequently, In th spreading of unsanitary material. Improvments in rural road and small bridges also belp to control soil and vegetation erosion. Two areas, sewerage and garbage disposal, hxowevor, will require special safeguards. Because of the project's civil works' cost liultation of US$ 500,000 equivalent, the project will not addrser directly the eavirormental b pact of the design and operation of, for exmple, sewerage treatsent plants or sanitary landfills. HBwever, for these and other investsents potentially hazardous to the nvirorment, all provincial Manuals would include the condition that any component, or larger infrastructure system of whlch the coponent is a; part, that does not fully comply with environmental protection legislation would not be eligible for financing under the project (para. 3.01(b)(Ix)). Furtherore, compliance with this condition would be specifically included in the terms of reference for ach province's semi-annual performance udit (para. 2.38). Risks and Saftruards 2.48 The main risk is lack of continuity in political comitment to the project at the provincial and/or municipal levels. This could occur with a sharp political change resulting in reductions in project counterpart funds and/or in effective management of the project by the Province's Project Zxecution Unit. 2.49 Because of the large demand for municipal infrastructure (Annex 1), and of the general lack of domestic, long-term financing, as well as the broad support for decentralization (with this support crossing political party lines), however, these risks are considered reasonable. Furthermore, the Loan Agreement would provide that If significant implementation delays occur in some provinces, then the initial total project cost allocation by Province would be adjusted to allocate more funds to those provinces with demonstrated greater implementation capacity (para. 2.20). ^ 23 - III. AGREEENTS REACHED AND RECOflENDUTIONS 3.01 During negotiations , assurances were obtalned, *a follows. (a) From the Central Goveriment tha t (1) if not all of the proceeds of th. Bank loan are withdrawn three years after the date of loan effoctiveness, then the remaining proceeds of the Bank loan will be reallocated (on an annual basis thereafter until the loan ts fully disbursed, or until the loan's closing date) according to each province's (1) willingness to continue in the project and (2) proportional participation in previous withdrawals of the proceeds of the loan (para. 2.20); (ii) it will open and maintain, in US dollars, a Special Account in a financial institution designated as Financial Intermediary on terms and conditions satisfactory to the Bank (para. 2.22); (iLL) all of the proceeds of the Bank loan will be onlent to the provinces with which the Bank has a Project Agreement (para. 2.22); (iv) it will fully assume the cross-currency exchange risk (para. 2.231 (v) it will designate and provide adequate support to a National Liaison Unit within the Secretariat of Housing and Environmental Management of the Ministry of Public Health and Social Affairs, and that this Unit will provide overall project implementation supervision, including consolidation of semi-annual performance and financial audits from each Province into project progress reports In support of Beak supervision (pares. 2.26-2.27); and (vi) it will have the Special Account audited according to standard, country-wide procedures agreed with the Bank (para. 2.38). (b) From the Provincial Governments that they will: Mi) establish and maintain Municipal Development Funds, the net proceeds of which will be, except as the Bank may otherwise agree, used exclusively for further onlending to municipalities at positive real interest rates (para. 2.09); ($$) assume the US dollar to Austral exchange risk, and pay a loan service charge of up to 12 of loan amounts disbursed (para. 2.23); ($ii) establish, adequately staff and otherwise support, In a manner satisfactory to the Bank, Project Execution Units during the entire project execution period (para. 2.27); - 24 - (Iv) submdt timely, annual lnvestsent program proposals (including =unicipal Investments and provincial technical asistance end training) for Bank review and approval (para. 2.28); (v) approve and carry out the program in accordance with the Project Operations Manual, and not amend this Manual without the lank's prior consent (para. 2.29); (vI) carry out procurement in accordance with Blnk guldelines (pares. 2.30-2.32); (vii) have the financial statements of the Projct Accounts in each province and participating municipality audited annually by external, indelpndent auditors acceptable to the Bank, and have *eml-annual evaluations of the compliance with all agreements entered into under the project provided through performance audits executed by external, independent auditors acceptable to the Bank (pares. 2.38-2.39); (viii) If significant deficiencies in provincial program implementation are identified in audits of the Project Account and/or the PEU, propose appropriate technical assistance ior the PEU as an obligatory part of the s*bsequent year's Investment program (para. 2.27), and if significant irregularities in municipal accounts andlor performance are identified in the periodic audits, make further project funding for physical investments in that municipality conditioned to a technical assistance program, acceptable to the Bank, focused on eliminating the observed ltregularities (para. 2.39); and (ix) carry out all environmental protection measures, as provided In the Project Operations Manual (para. 2.47); 3.02 The following would bo Conditions of Effectiveness: (a) subsidiary loan agreemnts between the Central Government and at least three Provinces (para. 2.22) that have (i) a legally established Municipal Development Fund (para. 2.09), (ii) legally established, adequately staffed and otherwise supported Project Execution Units (para. 2.27), (iii) formally approved Project Operations Manuals (para. 2.29), (iv) agreed to exempt all contracts for goods, works and services to be financed under the project from all legal and regulatory provisions limiting the international procurement of such goods, works and services (par&. 2.30), and (v) valid and binding Project Agreements with the Bank (para. 2.38); (b) execution of a Financial Agency Agreement between the Central Government and the Financial Intermediary (para. 2.22); and (c) formal designation, with adequate staffing and other support, of the National Liaison Unit of SVOA (para. 2.26). 25 - 3.03 The following would be Conditions of Diabursenent: (a) for all provinces not included in 3.02(a), *11 of the conditions noted therein; (b) in all provinces, audit reports acceptable to the Dank, and when Irregularities are reported, adequate measurs to correct then (puras. 2.27 and 2.39); and (e) in all provinces, all new subprojcts must bc Included in an Annual Investment Plan approved by the Bank (par&. 2.28). Recomondation 3 04 With the above assurance and conditions, the proposed project would be suitable for a Bank loan of US$ 120.0 million equivalent, to be repaid over a period of 15 years, including three years of grace, at the lank's standard variable Interest rate and fees. - 26 - Annex 1 Page 1 of 26 ARGENTINA CNICIPAL DEVELOMPMET PiOJECT Oeneral Demoarauhlc *nd Economic Trends and Cheraterristics of Partictuatina Provinces A. DEMOGRAPHIC TRENDS General Trends In Population and Hbousehold Growth 1. The longstanding trend of lncreasing urbanisation continued during the 1970s, with total urban population of Argentina growing at 2.42 per year, to 23.8 sillion In 1980, while the rural population declined from 5.1 to 4.7 million (see Table 1). Thus, the concentration of population in urban aresl increased from 702 in 1970 to 831 in 1980, and it is expected that thia trend will continue in the future. INDEC projects that urban population will grow by 5 million (2.5S per year) over the 1980190 period to 28.2 million In 1990. With thi growth, the share of total population living In urban areas will rise to almost 861 In 1990. 2. Ther ws, however, a striking reversal of another longstanding trend In the 1970st The share of the total population of the country living In the Buenos Aires Metropolitan Region (AMRA) dropped from 36.22 in 1970 to 35.92 in 1980. INDEC projecto a further decline to 35.22 by 1985. Over the 1970180 period, the urban agglomerations in the 500,000 to one million group increased their percontage of total population (from 7 to 12 percent),2 and the concentration of population In all urban agglomerations of over 500,000 group (including the AMBA) $ncreased from 431 to 472. However, this increase is due in large part to the entry, by 1980, of three new agglomerations in the 500,000 to 1 million population group (Gran Mendoza, Gran La Plata and Gran San Miguel de Tucuman). The number of cities In the 100 thousand to 500 thousand population range increased from 12 to 13 over the 1970180 period, but this group's participation in total population decreoased from 132 to 112. Finally, the cities of under 100,000 Inhabitants proved to be dynamic, lncreoasing their percentage of total population from 232 in 1970 to 252 in 1980, and their number from 596 to 693. 3. In sumary, there ws a slight tendency for decentralization of population growth away from the AMBA, with the concentration of Argentina's urban population residing In the region dropping from 45.92 in 1970 to 43.01 In 1980. This decentralization is undoubtedly related to the 11 The Instituto Nacional de Estadistica y Censos (INDEC) defines urban as a locality with 2,000 or more Inhabitants. A locality is a settlement with 10 or more housing units with separations of less than 100 meters mong the units. 21 There are a number of different table which present the evolution of the urban system with different class limits for the city sizes: The World Bank, Araentinas Public Sector Investment Review, World Bank, 1986, p. 94. Ministerio de Salud y Accion Social, Secretaria de Vivienda y Ordenamento Ambiental, Protrama Global de Financiamento de Infrasestructura v Eauiiamiento Urbanos, Argentina 1985. - 27 - Anx I Page 2 of 26 decentralisation of economic activity (discussed In Section I below). Hene, the much discuassd objective of spatial decentralisation of economic activitits and population fron the Metropolitan Region will be moving In the direction of existing demographic trends, not against thm.3 4. The rate of growth of total private households wa conasiderably higher then that of total population residing In these households over the 1960/80 period (2.41 versus 1.72, sea Table 2).4 The policy Implcatione of this phenoCmnon are quite important. If this trend continues, deogrephic demand for housing and urban services will continue to rise more rapidly than urban population per so. It appears however, that this will not be the cae, over the 1980190 period. The INDEC population projectiomn show that the size of the 20 to 29 age group will be only slightly larger in 1990 than in 1980. Thus, the total Aumber of households formed should be about tho same as during the 1970s. The reason for this is that fertility rates droppod rapidly during the 1950. and 1960s. This flattened out the age groups near the base of the age pyramid In 1980, meaning that the incraoe In young people entering the prime household formation ages will slow during the 1980s. However, as Table 3 shows, the growth of the 20 to 29 age group will pick up dramatically again in the 1990s, as the children born to fmilies formed in the 1970s begin to enter the 20 to 29 age group. This kind of wave action' is, of course, common in demographic growth. Population and Household Growth by Eliglibility for the Municiual Develovpmnt Proiect 5. Tablos 4 - 6 show population and household growth for 1970/90 by eligibility to participate In the Municipal Development Project. All municipalitIes5 In the Provinces of La Pampa and Nouquen are eligible. However, the rural area outside municipal boundaries is excluded in Neuquen. In the Province of Cordoba, only the Capital is ineligible, and only the Municipalities of Santa Fe and Rosario are not eligible in the Province of Santa Fe. In the Province of Buenos Aires, all of the 19 Partidos of the metropolitan region are ineligible, as are the localities of Mar del Plata in the Partido of General Pueyrredon and Bahia Blanca in the Partido of the same name. All of the other localities in these two partidos will be eligible, except the rural areas located outside of all of these localities. In the Partido of La Plate, the locality of La Plata is not eligible, but many of the outlying localities are eligible (see Table 7). The Attachment to this Annex provides further details on municipal eligibility. 31 This decentralization is discussed in two recent book_s D.F. Cavallo and J.A. Zapata, El Desafio Federal (Buenos Airest Planeta, 1986). M. Quadri Castillo, La Argentina Descentralizada (Buenos Aires, Tenms, 1986). 41 Unfortunately, it is not possible to compare the 1970 Census results on households with those of 1980, due to sampling problem with the 1970 census. 5/ The term municipalities is used to refer to all types of local govervment, including comisiones de fomento e comisiones vecinales. These latter entities are normally smaller comunlities. -28 - Annex 1 Page 3 of 26 6. Initially, it wa proposed that municipalties forming port of urban agglomerations crossing several municipal jurisdictions, as defined by INDZC, would be ineligible to participate In the project, in order to avoid problems resulting from the interdependence of infrastructure investment decisions when more than one jurisdiction is involved. Field examination showed, however, that the definition of urban agglomeration used by INDEC i, inadequate for the purposes of the projct. The INDEC definition is based on general functional interdependence and provides a suitable spatial unit for provision of statistical data. However, using the INDEC concept would have resulted in the exclusion of localities which are independent in terms of the kinds of infrastructure that the project would finance. It wa, therefore, decided to define conurbation in terms of this physical interdependence of infrastructure. In addition to being more relevant for the project, this definition would Increase its poverty impact, as many of the peripheral municipalities of urban agglomerations, that would have been excluded under INDEC's definition, would be eligible under the proposed operational definition of eligibility. 7. The population of the eligible municipalities grow from 5.2 million in 1970 to 6.0 millin in 1980 (1.5Z per year), and is projected to reach 7.0 million by 1990 (see Tables 4 and 5). In 1980, 21.7% of the total population of Argentina and 37.41 of that of the fiv, participating provinces lived in the eligible municipalities (see Table 6). The projected growth of 920 thousand in these eligible municipalities over the 1990s represents 18.61 of the total projected population increase for Argentina over this period. B. ECONOMIC ASPECTS 8. Table 8 provides an overview of the trends in economic growth over the 1970180 period for the five provincees. The rates of growth of Gross Regional Product (GRP) were very much higher in the four provinces outside of Buenos Aires. The slowest rate of growth was registered by the Federal Capital. Neuquen was the fastest growing, while Cordobs, La Pampa and Santa Fe grew at a rate approximately double that of the Province of Buenos Aires. Given these rates of growth, the relative participation of all of the participating provinces except Buenos Aires either remined constant or increased. Both the Province of Buenos Aires and the Federal Capital lost in the relative share of the total product, as they grew at rates much lower than the average for the country as a whole. 9. While the GRP per capita of the Province of Buenos Aires fell over the 1970/80 period, that of the other four provinces increased at rates of from 11.5% for La Pampa to 41.11 for Neuquen. Por this reason, all four of these provinces had ner capita GRP much higher then that of Buenos Aires in 1980. For example, the GRP per capita of Cordoba was 11.8Z higher than that of Buenos Aires; and Neuquen, 76 6Z higher. Note, however, that the nor capita GRP of the Federal Capital was 3 times higher than that of the Province of Buenos Aires and that it increased by over 18X during the decade. 10. Regional account data for all five provinces are not available after 1980, but the data on employment in manufacturing from the Economic Census of 1985 would lead to the conclusion that essentiallly the same - 2tA- Page 4 of 26 tendency tward dacentralisation of activity away from the AMA has prevailed. Argentina lost 10.01 of Its manufacturing jobs over the 1975185 petiod (Table 9). The Provinces of La Psapa and Neuquen actually gained manufacturing Jobs, as did the total of the other provinces outsdle of the five provinces and of the AMA. Santa Fe, Cordoba and the part of the Buenos Aires outside of the AMA lost manufacturing jobs at lower rates than the Metropolitan Region. In all, the ANA lost 167 thousand manufacturing jobs over the pertiod (-20.02). For the part of the country not included in the five provinces and the Federal Capital, the rate of job formtion was very high (23.42 or 54 thousand jobs). It is clear (Tables 9 and 10) that considerable decentralization of Industrial jobs occurred over the 1974185 period, as all areas outside of the AMU increased their shares of industrial employment. Evon so, 48.52 of all these jobs ware still located in the AMBA In 1985, down from 54.62 in 1974. C. INTER-GOVERNHMENTAL TRANSFERS AND THE CAPACITY TO PLAN CAPITAL INVESTMENTS AT THE MUNICIPAL LEVEL 11. Conceptually, there are two major reasons for inter-governmental fiscal transfers (a) because of legal andlor administrative attributes, one level of government has a comparative advantage in raising revenues over other lev1el of government, and it is therefore more efficient for it to collect a revenue and then share it with the other levels of government; and (b) a higher level of government is not satisfied with the distributional impact of the fiscal base status auo of other governments under its jurisdiction and centralizes the collection of certain revenues so that redistributional objectives may be achieved through the intergovernmental fiscal transfer system. When sub-national governments are expected to have a significant role in providing public services, public sector efficiency requires that sub-national governments receive fiscal transfers in a sufficient amount so that, combined with a reasonable sub-national government own-source fiscal effort, these governments may at least continue to maintain their current level of services. 12. In addition to quantity, predictability of inter-governmental transfers is important to public sector management efficiency. In Argentina, high variations in the annual amounts transferred to the municipalities from the Federal and provincial governments make it difficult for them to effectively plan their capital investments. The Public Sector Investment Review sums up the financial plight of the municipalities as follows: "Financially, municipalities constitute the weakest level of government in Argentina. Programming and multi-year investment planning at the municipal level is generally inadequate. ...Ad hoc federal sector allocations and direct transfers from the provinces increase the municipalities' financial dependency and create a high degree of uncertainty regarding the financing of physical infrastructure... Thus, the scheduling of medium or longer term works and capital investment progr mming becomes largely academic, and the actual works cost more than necessary because they are often undertaken on a piecemeal basis with frequent interruptions."6 13. In order to identify Improvements in the effectiveness of Argentina's inter-governmental fiscal tranfers system, it is necessary to 61 OD. cit. p. xix. -30 - Amx I Psge 5 of 26 Identify the trcanfer allocation criteri^, which may be classified as follow (a) ltralt d tlutLoD The taxes collected by the revenue sharing entity are divided accordin to the proportion of the total tax collected la eah jurisdictlon. In this case, the revenue sharer just acting as a tax collector for the other entity. (b) Fiscal Zff IcISe In this case, allocation of the shared funds to basod on soe ceasure of the level or the increaJe In tax receipts. for *x _l1, the distribution of the receipts of a given tax might be b ed on the total tax Oeffort' of the munilpality. (c) *baR ctgQ_btIn: Redistributional criteria Include population, surfce area, equal shares, the Inverse of population sise, indicators of underdevelopent, poverty and unestisfied basic needs. The higher the positive correlation of the criteria with a jurisdiction's total tax receipts, the lower the total redistributive Impact. To the extend that population slzs is nomally more highly and directly correlated with total tax receipts (rather then such criteria as surface area or equal shores) its reditributive impact will be lower. Redistributive impact Is highest where the criteria are lnversely correlated with total tax receipts, as is generally the case with poverty or underdevelopuent criteria. It Is, therefore, useful to divide the redistributive criteria Into two subgroups: proportional redistribution (population, surface area or equal shares) and comnensatori redistribution (inverse of population *ise, development gap, etc.). Federal Tranferro to the Provinces 14. The intitutional mechani_ms for tranferring resources from the Argentin Republic to the provinces are complex. for an overview, these mechanism may be classlfied by the following characteristics that can be suasisod in a 2 x 2 martrist 1) automatic or discretionary; 2) earmarked for a particular use *eg highways, housing, *tc) or free allocation by the lotal governments Free Allocation Earmarked Automatic a c Discretionary b d 15. Of most relevance to the proposed project are federal transfers over which provinces have allocation control, types a and b of the matrix: (a) Federal Revenue Sharing (Covarticipacion Federal) under Law 20.021 and Royalties for natural resource production. Royalties will not 71 Por a much more detailed analysis of these transfers, seet Alberto Porto, Anallsis del Impacto de los Anortes del Tesoro Nacional sobre l-s Finansas Provinciales (Buenos Aires: Consejo Federal de Inversiones, 1986). - 31 - Page 6 of 26 be delt with bhre becausw moe of the five provices under consideration disttibutes thm to municipalities (b) Discretionary Grants (Asortes del Tesoro Nacional) Because tboy are normlly not shared with unicipalitie and have relatively limited direct institutional impact at the local level, this Annex does not discuss the following two types (c) Automatic, *-rmark4 tranfors for the hlihg y dovelopment fund (Cosarttleticion Vial) and social security funds tranfcrred to pay pensions. (d) Discretionary, *rmarked trasfers for the Regional Developeant Fund (FDR), the Speciel Fund for Ilectrification of the Interior (FIPDU) and the Natinal Housing Fund (FrY!vI). Both of these two types are distributed back to the provinces using different criteria than types a and b. 16. Table 11 shows that the percentages of total Federal transfers to local goveriments through these four channels (abed) have varied tremendously over the years. For example, the total amount transferred via Revenu Sharing dropped from 66.22 In 1980 to only 20.11 In 1984, while the total from Discretionary Grants increased from only 52 in 1980 to 38.71 in 1984. In particular, Table 12 shows that, although Discretionary Grants hsve tended to rise to compensate drops in the Revenue Sharing and vice verts, the total amount going to the provinces has also been temporally unstable. For the 1980184 period, the annual amounts for Revenue Sharing ranged from US$1.0 billion to US$3.2 billion versus US$267 million to US$2.3 billion for Discretionary Grants. It should be noted that the percentage of the total transferred by Revenue Sharing dropped precipitously from 92.32 in 1980 to only 35.11 In 1984.8 Thus, the importance of Discretionary Grants had begun to rise even before the Revenue Sharing law expired at the end of 1984. The Criteria Used for Distribution of Federal Revenue Sharina 17. From 1973 through 1984, Federal Revenue Sharing to the proviwces was governed by a series of laws that defined not only the percentages of different federal taxes to be distributed, but also the critoris for distribution of these resources among the provinces.9 Immediately after the law 20.021 expired in 1984, *the central government continued.. as If the 8/ Porto discusses the resons for these fluctuations: op. cit. pp. 88 - 106. 91 This revenue sharing was originally regulated by Law 20.221 of 1973 which suffered various modifications over the years. However, the basic structure survived until 1984, when the prevailing legislation expired. For discussions of the evolution of this system, sees H. Nunes zinana and A. Porto, Distribucion de la Couarticiuacion Federal de Impuestos: Analysis y Alternatives (Buenos Aires: CFI, 1982). Jorge Macon, Perspectivas del Sistema do Coparticinacion (Buenos Aires, CFI, 1985). - 32 - Annex 1 Page 7 of 26 rcvenue-sharing arrangem.nts had been in effect."10 In both 1985 and 1986, Interim agreemente allocated rovenue sharing In accordance with what provinco had received In previous years.11 In April 1986, the executive brancb sent a proposal to Congress that would giva more weight to capensatory redistributional criteria.12 Under the existing and proposed systmo_, excise, inom, sales and inheritance taxes are shared. Under these Revenue Sharing lawa, the aount golng to a province Is determined In two basic steps. The first step (pre-primary nd prinary distribution) detemines the total Amount to be divided among the provincea in tho secondary distribution. In other words, prc-primary and primary distribution determine the 'sie of the pie" to be divided In the secondary distribution. At the primary step, tranfers to automatic (e.g. social security) and discretionary (e.g. FONAVI) earamrked funds are allocated. Then, the remaining amount Is divided so follows Federal government (46.52), th Municipal Government of the Federal Capital (1.8), Tierra del Fuego * Is1as d-l Atlantico Sur (0.21), the Regional Development Fund (3.02), and all of the Provinces (48.52). There is, therefore, some spatial redlstribution of resources in this primary stage. Between 1978 and 1981 the central government transferred responsibility for providing a number of important services to the provinces (hospitals, schools, sewr and water systems) without a compensating transfer of financial resources. In other words, there was a decentralization of responsibility but not of resources. The cost of providing these services would have required an Increase of about 222 of tho tctal amount shared.13 Provinces have, of course, been hard pressed to meet these now responsibility. 18. In the secondary distribution step, the total amount set aside for the provinces in the primary distribution step is then divided among them, according to the following criteria under the law which expired in 1984s 652 is divided in direct proportion to total population; 252, In inverse proportion to the level of development as measured by a number of indicators (Including housing quality, cars per capita nd educational level); 102, in Inverse proportion to population density for only those provinces with density levels under the national overage. All of the criteria used In this secondary distribution are, therefore, redistributive, and 352 ara of the compensatory type mentioned earlier, which have a very much higher redistributive impact.14 Table 13 permits comparlson of the percentage distribution of Revenue Sharing funds and Discretionary Grants with the distribution of population (which, as discussed earlier, is Itself a redistributive criteria). If secondary distribution were done only according to population, Buenos Aires would 10, A. H. Petrel, "Fiscal Relations Between the Central and Provincial Governments of Argentina," World Bank, xerox, 1987, p. 8. 11/ For more details, see Petrei, oD. cit., p.9. 121 ibid., pp.10-11. 131 Ibid., pp. 6-7. 141 The proposed law sent to the congress would increase the weight on the development gap criteria by 10 percent points. Petrei, op. cit., pp. 10-11. -33- Amex I Page 8 of 26 have received 43.52 of the total. Howover, under the Revenue Sharing Iawo| It received only 28.3Z (see colume 2). Buenos Alres faired even worse with the Discretionary Grants, receiving only 17.52 of the total In 1985. The ladies of concentration of this so table show that both Neuquen and La Pam received about twice as such as they would have If the resourecs had bee tranforred on the basis of population. In umJary, the criteria used In Federal Revenue Sharin are exclusively rodistributive in character and the redistributive Impact very high, even when compared with population distribution. Proposed ligislation would lncrease this redistributive mpact. Rovenue Sharint fror Provinces to Hunlcitnlities 19. Table 14 presents a summary of the percentages used In the prilary distribution of Federal Revenue Sharing funds and provincial taxes for distribution to the municipalitles. The provinces vary greatly not only in what taxs they share, but also In the percentages that they allocate for distribution among the different municipalties. For example, the Provinces of Buenos Aires and La Pampa set asIde a constant percentage of almost all of its taxes and Federal revenue sharing to be divided, while the Province of Santa Fe allocates widely ranging proportions of the different provincial taxes for secondary distribution among the Municipalities. 20. Table 15 shows that the total amounts transferred by the five provinces to municipal government varied widely for the years 1970, 1979, 1973 and 1984. For example, the Index showes that the total amount transferred to Cordoba in 1983 was 20.32 higher than 1984 and about the same in 1984 as In 1979. For Neuquen, the amount transferred was over 302 higher than 1984 in 1983, but about 272 lower than 1984 In 1979. Clearly, such temporal instability In the transfers makes effective planning of capital investments difficult. The cause of these fluctuations are the variations iln the amounts received through Federal Revenue Sharing and other transfers, as well as In provincial tax rocelpts. The Impact of this temporal instability at the municipal level depends on the degree of provincial dependence on revenues derived from Federal sources (including royalties for extraction of natural resource). This dependence varies greatly. In 1985, for *xample, only 37.52 of the total revenue of the Province of Buenos Alres was from Federal sources, versus 56.22 for Santa Fe, 61.22 for Cordoba, 77.62 for La Pampa and 82.32 for Neuquen.15 21. Table 16 shows that the criteria used In secondary distribution of resources to the municipalities are predominately redistributive In nature, except for the Provinces of La Pampa and Neuquen. Even for these two provinces, 252 of secondary distribution in La Pempa and 402 in Neuquen is done using redistributive criteria. For Buenos Aires, the criteria used are exclusively redistributive, and for both Santa Fe and Cordoba, about three quarters of the secondary distribution is done using redistributive criteria. 22. In sumary, the Federal Revenue Sharing system uses exclusively redistributive criteria. For the provincial systems, the criteria vary 15/ See Table 4 in Petrel, ov. cit. - 34 - Annex 1 Page 9 of 26 considerably, but all use some redistributive criteria. Because of thc fluctuations Ln Federal Revenue Shoring and In provincial tax receipts, the amounts transfrrced annually to municipalitie by the provinces have varied greatly over tim. The Provincial and Muneciual Shres of Total Public Ixuenditures 23. During the 1960c and 1970., the fiscal significanca of Argentina's 23 provinces and territories grew substantially. However, *lthough current and capital expenditures of th. provinces increased from US$4.5 billion In 1970 to US$7.1 billion In 1980, the tendency in the 1980. has been a decline en real terms to US$5.8 billion In 1985 (-23.01).16 The 1980a have also boen charactorised by tmporal Instability of expenditures at the provincial level with total expenditures ranging from US$6.8 billion in 1981 to US$4.4 billion in 1981. The percentages of total public expenditures going to provincial goverrments also reflect this instability, ranging from 15.42 In 1982 to 19.01 ln 1984. In terns of GDP, the percentage going to the provinces has varied considerably during the 1980s ranging from 8.1S In 1980 to 10.01 in 1983, with the general tendency betag upward, reflecting the trend in total public expenditures, in which this percentage rose from 38.6Z in 1970 to 52.12 in 1985.17 24. The municipalities fared much worse than the provincial governments during the 1980s, showing absolute and rolative declines in total current and capital expenditures. Total municipal expenditures dropped by nearly US$1 billion over the 1980/85 period, and the percentage of total public expenditures, from 4.61 to 3.01. As a percentage of GDP, this meant a decline from 2.21 in 1980 to 1.62 in 1985 (to almost the 1970 level, 1.51). Temporal listability was also high during the 1980185 period with total expenditures ranging from US$900 million to US$1.9 billion. 25. Municipal governments have relatively low access to credit compared to the Central Administration or Provinces. The monthly average lent to munielpalities In 1986 was US$48 million while the Central Government and Provincial Administrations received over four times this amount Furthermore, while the average monthly amount lent to the municipalities in 1986 was about equal to that of 1984, the average going to the provinces increased by almost five times 18 The mayors inverviewed during project preparation were unanimous in stating that they had very little access to credit and what was available was for only very sbort terms (normally less than two years and mostly for equipment). For this reason, municipalities tend to have considerable unused borrowing capacity. 16/ See Table 17. 17/ The very low and negative annual rates of growth of the GDP explain part of this incrasee. 181 For a more complete analysis, seet A.H. Petrei, *Fiscal Relations Between the Central and Provincial Government of Argentinal, Preliminary version of a paper prepared for the World Bank, xerox, 1987, Tables 8, 9 and 10. - 35 - Page 10 of 26 $TTACDINTN LIO!ISILITY FOR PRIOJIC? PARTICIPATION BY PROVINCX Prowinc of Buenos Aires All Partidos excepts - the 19 Partidos of Gran Buenos Alrces. - the Partido of La Plata with the xception of the following localities which are considered eligible: Arturo Sougn, Govina, Jose Eern and, Abato, Lissndro O0ao, Melchor Roiuro, City 5*11 and Villa Ilisa. - the localities of Bhla Blanc and Mar de Plata. All of the other localities of the Partidos of Bahla Blanca and General Pusyrredon would be eligible. Province of Cordobas All Municipalities except tho Capital would be eligible. Province of La Pmp-a All Municipalities would be eligible. Province of Neuquens All Municipalities would be eligible, but th rural areas outside of municipal boundaries would not be eligible. Province of Santa Fe: All Municipalities except Santa Fe and Rosario would be eligible. - 36 - Annex1 ARGENTINA Page 11 of 26 MUNICIPAL DEVELOPMENT PROJECT Table I Total. Urban and Rural Population: 1970/80 and Projection to 1990 Years Millions of inhabitants X of Total Population (a) Total Urban Rural Total Urban Rural 1970 23.4 18.3 5.1 100.0 78.4 21.6 1980 27.9 23.2 4.7 100.0 83.0 17.0 1990 32.9 28.2 4.7 180.8 85.8 14.2 Annual Rate of Growth 1970/80 1.8 2.4 -9.6 1980/90 1.7 2.0 -0.1 SOURCE: INDEC. CENSO NACIONAL DE POBLACION Y VIVIENDA: 1980. SERIE 0, POBLACION. INDEC. PROYECCION DE POBLACION 1970-2025. Table 2 Total Private Households and the Population Residino in Them: 1960/80
Группа Всемирного банка · Staff Appraisal Report
Argentina - Municipal Development Project
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст
Основные сведения
Организация
Группа Всемирного банка
Тип документа
Staff Appraisal Report
Страна
Аргентина
Источник
Всемирный банк