Document of The World Bank

Transcription

1 Document of The World Bank Public Disclosure Authorized Report No AR Public Disclosure Authorized STAFF APPRAISAL REPORT Public Disclosure Authorized ARGENTINA SECOND MUNICIPAL DEVELOPMENT PROJECT MARCH 6, 1995 Public Disclosure Authorized Environment and Urban Development Operations Division Country Department I Latin America and the Caribbean Regional Office

2 CURRENCY EQUIVALENTS Currency Unit - Peso (Arg$) EXCHANGE RATE Arg$1 = US$1 WEIGHTS AND MEASURES Metric System FISCAL YEAR January 1 - December 31 ABBREVIATIONS AND ACRONYMS CCU - Central Coordinating Unit (Unidad de Coordinaci6n Central) EA - Environmental Assessment GDP - Gross Domestic Product 1DB - Inter-American Development Bank IERR - Internal Economic Rate of Return INDEC - National Institute for Statistics and Census (Insntuto Nacional de Estadtsticas y Censos) IS - Institutional Strengthening MDF - Municipal Development Fund PCR - Project Completion Report PEU - Provincial Executing Unit POM - Project Operations Manual PPAR - Project Performance Audit Report PRAM - Brazil Parana Market Towns Project SDS - Secretariat of Social Development of the Presidency (Secretarfa de Desarrollo Social de la Presidencia) SOE - Statement of Expenditures SV - Under Secretariat of Housing (Sub-Secretarfa de Vivienda) VAT - Value-Added Tax

3 ARGENTINA SECOND MUNICIPAL DEVELOPMENT PROJECT TABLE OF CONTENTS LOAN AND PROJECT SUMMARY... PageNo. i I. THE SECTOR... 1 Background... 1 Government Structure for Provision of Public Services... 2 Sector Development Issues and Objectives... 3 Public Sector Financing and Intergovemmental Fiscal Relations The Efficiency and Equity of the Intergovernmental Transfer System... Provincial Revenues Provincial to Municipal Revenue Transfers... 6 Municipal Own-Source Revenues... 7 Sector Strategy and Rationale for Bank Involvement... 7 Lessons from Previous Bank Involvement in the Sector... 8 II. THE PROJECT Project Origins Project Objectives Project Components Eligibility Criteria Provincial Eligibility Criteria Municipal Eligibility Criteria Subproject Eligibility Criteria Project Costs Allocation of Loan Proceeds by Province Financing Plan Implementation Arrangements Procurement Disbursement Accounts and Audits Financial Impact Economic Impact Poverty Impact Environmental Impact Risks and Safeguards HI. AGREEMENTS REACHED AND RECOMMENDATION Assurances Conditions of Effectiveness Condition of Disbutrse nt Recommendation... 32

4 ANNEXES A. Previous Bank Experience in the Sector B. Population Growth and Distribution C. Demand for Municipal Infrastructure D. Intergovernmental Fiscal Relations E. Initial Allocation and Commitment Fees by Province F. Economic Analysis G. Implementation and Supervision Arrangements H. Disbursement Schedule I. Selected Documents and Data Available in Project File MAP IBRD No LIST OF TABLES Table 2.1: Estimated Program Costs Table 2.2: Initial Allocation of Loan Proceeds by Province Table 2.3: Financing Plan Table 2.4: Procurement Method by Category Table 2.5: Procurement Prior Review and SOE Thresholds This report is based on the findings of a preappraisal mission to Argentina from June 21 to July 8, 1994 and was drafted in the field. The mission comprised: Messrs/Mmes. Mario Rothschild (Mission Leader and Sr. Urban Specialist); David Vetter (Sr. Urban Specialist); Miguel Mercado-Diaz (Operations Analyst); Craig Leisher (Writer and Editor); and Carlos Ramirez, Gunnar Stahre, and Cecilia Zanetta (Consultants). Messrs. Asif Faiz, Orville Grimes and Gobind T. Nankani are respectively the managing Division Chief, Projects Adviser, and Department Director. The peer reviewers were Messrs. Yoshine Uchimura, EC4HM, James Hicks, TWURD and William Dillinger, LAlEU.

5 ARGENTINA SECOND MUNICIPAL DEVELOPMENT PROJECT LOAN AND PROJECT SUMMARY Borrower: Implementing Agencies: Beneficiaries: Poverty: Amount: Terms: Commitment Fee: Onlending Terms: Argentine Republic. Provinces of Buenos Aires, Catamarca, C6rdoba, Corrientes, Formosa, La Rioja, La Pampa, Mendoza, Misiones, Neuquen, Santa Cruz, Santa Fe and Tierra del Fuego; and the eligible municipalities in these provinces. Approximately 900 municipalities in 13 provinces. Not applicable. US$210 million equivalent (including up to US$21 million in retroactive financing). Repayment in 15 years at the Bank's standard variable interest rate, including 5 years of grace percent on undisbursed loan balances, beginning 60 days after signing, less any waiver. All proceeds of the loan would be onlent to provincial governments from the Central Government under the same terms and conditions as the Bank loan. The provinces would assume the cross-currency exchange risk and would pay a onetime service charge of up to 2.5 percent of the loan amount to cover the costs of a Central Coordinating Unit. With the exception of amounts up to 5 percent of the loan, which may be used by the provinces to finance technical assistance, all proceeds of the Bank loan would be onlent from provinces to municipalities in US dollars and or. harder terms, with a maximum of 10 years repayment, including up to one year of grace, and a spread of at least 1.5 percentage points over the interest rate paid by provinces to the Central Government. Plnancing Plan: See para Rate of Return: 19.7% for an estimated 54 percent of total project cost.

6 - ii - Staff Appraisal Report: No AR, dated March 6, Map: IBRD No

7 ARGENTINA SECOND MUNICIPAL DEVELOPMENT PROJECT I. THE SECTOR A. Background 1.1 The move toward a more efficient, equitable and fiscally responsible federalism is a key element in Argentina's macroeconomic strategy. Following a policy of decentralization, Argentine public services are to be provided at the most decentralized level consistent with providing the service efficiently. In this context, provinces have decentralized or are planning to decentralize services to municipalities if these municipalities can provide them more efficiently. Current legislation establishes municipal responsibility for a number of services, including the paving and maintenance of streets and local roads, water and sewerage, refuse collection, and some health, education and other community facilities (para. 1.9). 1.2 The municipal public sector is large and growing in Argentina, with expenditures of about 3.5% of GDP. Transfers from provincial to municipal governments have doubled, increasing from US$1.4 billion in 1991 to US$2.8 billion in These transfers represented about 1% of national GDP and 11% of total provincial spending in 1993, ranking second after salaries. Population Growth and Distribution 1.3 Argentina is one of the most urbanized countries in Latin America, with the unique characteristic that urbanization was well advanced before World War I. In 1991, 87% of the total population lived in urban areas of 2,000 or more inhabitants. Despite its high level of urbanization, Argentina is still experiencing significant growth in its urban population, which grew from 23.2 million in 1980 to 28.4 million in 1991 (1.85% per year). Conversely, the rural population has decreased at a rate of 0.82% per year over the same period (Annex B). 1.4 The 13 provinces participating in the proposed project represent 70% of the total population. These provinces can be divided by size into two major groups: (a) four highly-populated provinces (Buenos Aires, C6rdoba, Mendoza and Santa Fe) with 19.6 million inhabitants (60% of the total Argentine population); and (b) nine less-populated provinces (Catamarca, Corrientes, Formosa, La Pampa, La Rioja, Misiones, Neuquen, Santa Cruz and Tierra del Fuego) with 3.3 million inhabitants (only 10% of the total population). 1.5 The extreme diversity of the provinces of Argentina in terms of population can be seen by comparing the largest province with one of the

8 smallest: Buenos Aires and Santa Cruz. The population of the Province of Buenos Aires (12.6 million in 1991) is about equal to that of Chile, but lives in an area about the size of Italy with a density of 41 people per square kilometer. On the other hand, in the Province of Santa Cruz 160,000 inhabitants live in an area the size of the United Kingdom, giving the province a population density of less than one person per square kilometer. Although the percentage of poor households is higher in the provinces with smaller populations, over half of the total number of poor households are in the three largest provinces - Buenos Aires, C6rdoba and Santa Fe - and 42% of the Argentine population lives in the capitals of these provinces (the Buenos Aires, C6rdoba and Santa Fe metropolitan areas). The distribution of municipalities follows a similar pattern, with 68% of all municipalities covered by the proposed project located in the same three provinces. Unmet Demand for Basic Municipal Infrastructure 1.6 During the years of Argentine macroeconomic instability, investment in basic municipal infrastructure lagged far behind the needs of the population. For example, the number of households without paved streets in the capitals of Buenos Aires (13 million inhabitants in 1991) and Santa Cruz (145,000 inhabitants) has increased by 25% and 15% respectively over the last decade. 1.7 The potential demand for basic municipal infrastructure was estimated from a representative sample survey. In the 13 provinces, total potential demand is approximately US$1.3 billion, of which 90% are subprojects under US$1.5 million. Most of the subprojects identified are in the street paving and sanitation sectors, which account for 31 % (US$406 million) and 21% (US$281 million), respectively. B. Government Structure for Provision of Public Services 1.8 Argentina has a federal system of government with three levels: central, provincial and municipal. Until the adoption of the Federal Constitution of 1853, provinces were sovereign in their own right. Through this Constitution, provinces surrendered wide-ranging powers to the Central Government, but all powers not explicitly relinquished to the center remain with the provinces. Argentina's 23 provinces are autonomous entities governed by their own constitutions. Although the Federal Constitution establishes that municipalities should be independent from other levels of government, each provincial constitution provides the basic framework for its municipal governments and requires a Municipal Code (Ley Orgdnica Municipal) that establishes specific municipal responsibilities for service provision and sources of revenue.

9 Since 1991, the Central Government has assigned responsibility for provision of some additional services to provincial governments (e.g., primary and secondary education, and health care). In some cases, the provinces have, in turn, passed on these and other responsibilities to municipalities. Responsibilities for service provision may be illustrated as follows: exclusively Central Government - defense, foreign affairs, inter-province transportation and trade regulation, mail and telecommunications; Central and provincial govermnents - higher education, preventive health, justice, security, economic development, major passenger and cargo terminals, housing, electric and gas energy; provincial and municipal governments - elementary and secondary education, health care, water and sewerage, regional and local roads, and fire control; and predominantly municipal governments - solid waste collection and disposal, local streets and drainage, parks, markets, cemeteries, and land use planning and control As Argentina's decentralization program progresses, there is a continuing need to confirm that the assignment of responsibilities for the provision of services is clearly and unambiguously defined among the different levels of government. Currently, the central to provincial relationship has relatively welldefined areas of responsibility for the provision of services. It is the provincial to municipal relationship where the responsibilities could be more ambiguous. C. Sector Development Issues and Objectives 1.11 Issues and objectives in the urban sector are primarily those related to 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 transfer, to the maximum extent possible, the full financing burden for public investments and recurrent costs to their direct beneficiaries. The central and provincial govemments have limited potential to recover costs directly from beneficiaries. At the municipal government level, however, there is substantial potential, frequently under-utilized, for direct cost recovery from beneficiaries. Furthermore, as the Central Government's decentralization policy proceeds, this potential should increase. Instruments for improved municipal direct cost recovery include betterment levies, connection fees, user charges, concessions and rents. To the extent that these direct cost recovery mechanisms can be improved and expanded, municipal governments would contribute substantially to national objectives of resource mobilization in a non-deficit and noninflationary manner The resource mobilization objective is closely linked to the objective of improved public sector management and efficiency. To effectively mobilize

10 resources, municipal govemments must enhance their capacity to plan, program and budget for capital improvements programs, as well as for operation and maintenance of facilities. Effective decentralization and improvement of public sector management at the subnational level depend greatly on the clarification of the respective expenditure responsibilities of provinces and municipalities as well as assignment of revenues to meet these responsibilities. There are at least three key issues for successful implementation of these improvements: (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 responsibility for the resources transferred; and (c) how to encourage efficiency reforms in provincial and local governments in areas such as planning, budgeting and financial programming All of the above are linked to the issue of accountability. The main barrier to improving provincial fiscal performance is the lack of accountability at both the provincial and municipal levels. This stems from the strong central financing of provincial expenditures and from a high level of provincial financing of municipal expenditures. The result is a lack of incentive for provincial and municipal authorities to confront their citizens with the costs of their preferences. The lack of accountability undermines incentives for provincial and municipal fiscal balance and for provincial and municipal expenditure efficiency. D. Public Sector Financing and Intergovernmental Fiscal Relations 1.14 Both provincial and municipal governments in Argentina are highly dependent upon transfers from higher govemmental levels. The central government transfers resources to the provinces through automatic revenue sharing and discretionary grants. The provinces, in turn, are required by the national revenue law to transfer part of the resources down to their municipalities via automatic revenue sharing. In addition, most provinces also transfer shares of their own taxes to the municipalities via automatic revenue sharing, and may provide discretionary grants to municipalities. Thus, in the Argentine system, provincial revenues in large part determine municipal revenues. The Efficiency and Equity of the Intergovernmental Transfer System 1.15 With regard to the impact of national to provincial transfers, a recent Bank study concluded that the regional redistribution of income probably has been excessive in Argentina. Although data are lacking to evaluate the efficiency of provincial expenditure programs (e.g., education, health, and transport), there is evidence that serious inefficiencies exist regarding types of expenditure, especially for salaries.

11 1.16 The issue is, then, less the amount of provincial revenues than in how efficiently existing revenues are allocated. To address the issue of expenditure efficiency, the criterion of the accountability at the provincial level must be examined. While the more developed provinces must confront their citizens with over half the cost of financing their expenditures, provincial authorities of the less developed provinces have a much lower political cost, as more than 75% (90% in the case of underdeveloped provinces) of the fiscal burden for financing their expenditures falls on the payers of central government taxes. The combination of low own-source fiscal effort and of large, generally untied (block) fiscal transfers provides little accountability for many provincial officials, either to their own citizens or to the payers of national taxes. This lack of accountability provides little incentive for an efficient allocation of resources in most provinces. As shown below (para. 1.21), municipal dependence on provincial transfers is also relatively high, creating this same lack of accountability on the part of municipal officials. Provincial Revenues 1.17 Due to the impact of macroeconomic stability on national and provincial revenues, total provincial revenues (including both own-source and transfers from the national govemment) jumped from US$14.3 billion in 1990 to US$24.7 billion in In 1993, national to provincial automatic revenue sharing under the national law (hereafter, coparticipaci6n) reached US$10.4 billion (with a total of US$14.3 billion for all automatic transfers and discretionary grants from the national government), and provincial own-source current revenues were US$10.4 billion. While of diminishing importance, discretionary transfers from the national government to finance provincial deficits and investments have been particularly insidious in rewarding poor fiscal performers among the provinces and have significantly distorted resource allocation. Although the Central Government has explicitly sought to diminish their size, discretionary transfers still represented about 8% of total provincial revenues in 1992 versus 19% in 1987 and remain important sources of revenue for many smaller provinces National transfers financed on average 59% of total provincial spending in 1992, ranging from a high of 92% of total expenditures in La Rioja to a low of 50% of spending in Buenos Aires. As mentioned above, this high provincial dependence on transfers of federal revenues has direct and important impacts on municipal revenues, as national coparticipaci6n resources are automatically shared with municipalities and can also be used for discretionary transfers to the municipalities The prevailing national coparticipacion law allocates to the provinces 57.0% of the most important taxes collected, such as the value-added tax

12 - 6 - (VAT), income and asset taxes, and excise and fuel taxes. In August 1993, a floor of US$745 million per month was set on these transfers. Secondary distribution of coparticipacion revenues among provinces, codified by the coparticipaci6n law, favors sparsely populated and less-developed provinces at the expense of the more developed and densely populated provinces which have a higher concentration of the nation's households with unsatisfied basic needs. Provincial to Municipal Revenue Transfers 1.20 Intergovernmental fiscal relationships between provinces and municipalities are similar to those between the central and provincial governments, with each provincial constitution providing for specific arrangements. Provincial to municipal transfers may be automatic or discretionary Between 1983 and 1990 total transfers from the Central Government to provinces fell from US$3.6 billion to US$3.1 billion, but the share of transfers to the municipalities doubled from 23% to 50% for the period (from 8% to 15% of total expenditures). The increase in transfers can be attributed to the expanding employment roles of the municipalities and a decline in their own tax efforts. With the recent increase in federal transfers to the provinces, transfers to municipalities have also increased. These transfers rose, from US$1.3 billion in 1990 to US$2.4 billion in 1992, frequently without corresponding increases in responsibilities Automatic Revenue Sharing. The provincial to municipal coparnicipaci6n laws define the primary distribution of resources from national revenue sharing and provincial taxes to be shared with the municipalities and then define the criteria for distribution among the municipalities (i.e., secondary distribution). For the 13 Bank provinces, the primary provincial to municipal distribution of total national automatic revenue sharing in 1992 ranged from 5% in Corrientes to 30% in Tierra del Fuego. For provincial taxes, the percentage going to the municipalities ranged from 14% in Buenos Aires and Mendoza to 80% in Neuquen (Annex D) Secondary distribution in provincial to municipal revenue sharing can be classified as redistributional (e.g., criteria such as population, land surface, equal share, density, and basic needs) or related to fiscal effort (e.g., municipal tax effort and fiscal health). In the 13 provinces to be assisted by the Bank loan, the 1992 percentages show that the amounts allocated for fiscal effort vary greatly, ranging from zero in Corrientes to 64.3% in La Pampa (Annex D) Discretionary Transfers. Discretionary provincial to municipal grants can cover budget deficits or finance investments. In a recent sample for the

13 - 7 - project's provinces, total discretionary grants from the provinces in 1992 exceeded total own-source and coparticipaci6n revenues by 10% in over onequarter of the municipalities surveyed. It is safe to assume that the greater the provincial discretionary grants, the less likely it will be that municipalities will confront their constituents with the costs of their investment decisions. Municipal Own-Source Revenues 1.25 The primary municipal own-source revenues are: (a) user charges for public services (frequently tied to the property tax) such as refuse collection and disposal, and street maintenance; and (b) betterment levies for public works that directly enhance property values. Although the dependence on provincial automatic transfers versus own-source revenues varies widely, about half of the municipalities in a representative sample obtained more than 60% of their total current revenues (excluding discretionary grants) from own sources. Only about 11% obtained more than 60% from automatic revenue sharing. Thus, though the level of municipal dependence on provincial transfers is relatively high, municipalities do have their own sources of revenues (Annex D). E. Sector Strategy and Rationale for Bank Involvement 1.26 The Bank has been, and remains, a strong supporter of the Government of Argentina's ambitious public sector reform program. This project would assist in extending the fiscal and administrative reforms at the national level to the municipal public sector. By increasing the resource base and service delivery efficiency of municipal governments, the project would support increased public sector efficiency (by confronting final beneficiaries with the costs of their decisions, thereby increasing accountability) and resource mobilization (for counterpart funds during implementation and through cost recovery). A more efficient municipal public sector would be able to provide more effective infrastructure to support productive activities, meet the basic needs of the population, and improve environmental quality. By advancing efforts to rebuild the infrastructure base and consolidate macroeconomic reforms, the proposed project fully supports the Country Assistance Strategy discussed by the Board on March 1, Specifically, municipal capital improvements programs that offer adequate cost recovery, primarily from the program's direct beneficiaries, explicitly support these objectives. Greater cost recovery from these beneficiaries would increase the accountability of municipalities and, therefore, the efficiency of resource allocation. By promoting fiscal discipline and an appropriate balance between expenditure responsibility and revenue authority by level of government, the project would support resource mobilization in a financially responsible manner. In addition, as in the Provincial Adjustment

14 - 8 - Loan (proposed for FY95), the project would support the design and implementation of provincial to municipal revenue sharing systems that are more equitable and promote improvements in municipal fiscal performance By making municipalities more fiscally autonomous and thereby reducing pressures for provincial transfers to them, the proposed project would also contribute to the improvement of the fiscal performance of the provinces, the main goal of both the above-mentioned adjustment loan and the Provincial Development Project (US$200 million, FY91). F. Lessons from Previous Bank Involvement in the Sector 1.29 The proposed project is similar to the newer style of municipal development projects that stress the improvement of efficiency and fiscal discipline of the municipal public sector, such as in the first Argentina (Loan 2920-AR) and Ecuador (Loan 3285-EC) municipal development operations. As these municipal finance-oriented operations are relatively new, there are only a few Project Completion Reports (PCRs) and Project Performance Audit Reports (PPARs) available on this type of operation. Also, a recent report issued by OED (7Wenry Years of Lending for Urban Development, ) concentrates its analysis primarily on the previous traditional type of urban projects, reviewing the relationship between these projects and the key items now identified by the Bank for its urban policy agenda: improving the urban environment, poverty reduction, economic growth and increased efficiency of infrastructure services. However, the report also comments that "from the outcomes of completed operations there is evidence of impacts through real sector, fiscal, and financial links... but there is scope for improvements in 'operationalizing' these links." 1.30 Of the projects reviewed, the experience of the Parand Market Towns Project (PRAM) is particularly relevant for the proposed new project. The PPAR for this project concludes: PRAM's success can be attributed to a number of factors: (a) it was based on an existing local initiative that was able to maintain strong political support even when control of the state government shifted from one party to another; (b) most subprojects were small and of relatively simple design and easy to implement; (c) it was administered by a competent and motivated team and was able to utilize the decentralized structure of the existing state municipal assistance agency; (d) it was able to meet the priority demands of a large number of municipalities, providing it with additional political support; (e) as a multi-town program riks of delayed or non-execution were spread over a large number of sub-borrowers; and (f) its location in a state having a tradition of good public administration provided a propitious institutional setting for project execution (Report No , paras ).

15 The later OED study (Thvenry Years of Lending) was a much more extensive study, reviewing 115 completed urban projects in 52 countries. Many of its conclusions repeat and confirm earlier observations. Specifically, it recommends the following lines of action: (a) tapping the potential of real, fiscal, and financial linkages of urban projects to enhance their impact; (b) strengthening the congruence of objectives and design; (c) securing project ownership (by the beneficiaries); and (d) improving monitoring and evaluation systems A recent PCR for a municipal development project in Honduras (Loan 2583-HO) concludes with a number of useful lessons: (a) the need for continuing close coordination among donor agencies; (b) the need to address the system of incentives within which local governments operate; (c) the need to provide institutional strengthening programs within an appropriate environment and sufficient time horizon; (d) the desirability of incorporating private capital markets to mobilize and allocate savings for municipalities; (e) having welldesigned decentralization policies; and (f) the need for close Bank supervision in the case of programs supporting institutional reform. Hurst Municipal Development Project 1.33 Particularly relevant for the proposed project are the lessons learned from the success of the Argentina First Municipal Development Project (Loan 2920-AR). In terms of investments, the success has been due primarily to the following factors: (a) the concentration on small, simple and easy to implement investments; (b) the coverage of a large number of interested municipalities; (c) the maintenance of very competent local executing units; and (d) the flexibility of shifting and reallocating loan funds among competing provinces A related factor in the success of this project has been the limited role assigned to the federal level, where a small but effective coordinating offlce is in charge of routine contacts with the Bank and some project review functions. This has succeeded in avoiding the delays sometimes created by a large, controlling central unit, and greatly contributed to the efficient implementation of this project. In addition, this project benefitted by having only a few and simple "pass - no pass" fiscal conditions. These were easy to quantify, easy for the municipalities to understand, and simple for the Bank to review and approve. Regarding subproject cost recovery, the experience of the first Municipal Development Project (Loan 2920-AR) was satisfactory. Ninety percent of the loan proceeds had cost recovery greater than 50%, and 63% of loan proceeds had cost recovery greater than 65%.

16 - 10- II. THE PROJECT A. Project Origins 2.1 The proposed operation would be the second of its kind in Argentina. It builds on the operational experience of the first Argentina Municipal Development Project (Loan 2920-AR) and would complement - at the municipal level - the fiscal objectives of the Provincial Development Project (Loan 3280-AR). The first Municipal Development Project became effective in March of 1989 and covered five provinces. Despite a slow start caused by the macro conditions prevailing at the time, the project is expected to be completed on schedule, as all available funds have already been fully committed. Many more eligible investments in the five provinces have been identified, but cannot be considered due to lack of funds. Similar opportunities have been identified in the other provinces of Argentina. The Central Government has therefore asked the Bank to consider this new operation. A parallel operation was approved by the Inter-American Development Bank (IDB) in late To avoid implementation difficulties, the Central Govemment has asked each bank to work in a different group of provinces for their respective new operations, except that the two banks would work together in the four largest provinces, each financing a different group of investments according to size (the World Bank financing all municipal investments up to US$2.0 million equivalent and the IDB financing the rest). B. Project Objectives 2.2 The main objective of the project is to contribute to more effective public sector management at the provincial and municipal levels through improved financing mechanisms for municipal investments. It would also help attain more effective fiscal federalism by strengthening the municipalities' capacity to assume the responsibilities being transferred to them. The specific objectives are to: (a) (b) mobilize external and internal resources in a non-deficit and noninflationary way in order to finance justified municipal investments - particularly those contributing to productive activities; strengthen municipalities' capacity to plan, finance and execute, in an efficient manner, cost-effective capital investment programs, thus assisting them to develop the capacity to eventually finance their investment needs through the private capital markets; and

17 (c) strengthen the institutional capacity of municipalities and provinces to manage their resources more effectively and to increase the efficiency of their systems for delivery of services. C. Project Components 2.3 The project would include the following components. (a) (b) (c) Municipal infrastructure (81% of total project costs), including construction and rehabilitation of public infrastructure (e.g., road paving and related works, public lighting and sanitation), community facilities (e.g., markets, and bus terminals), equipment for other municipal services (e.g., for road maintenance, refuse collection and disposal). Institutional strengthening (6% of total project costs), including improvements in financial management, information systems, accounting procedures, cadasters, maintenance procedures, and analyses to identify key policy issues and make recommendations for improving the assignment of revenues and responsibilities between the provincial and municipal levels of government. Although provinces could use up to 5 % of the amount of the loan (US$10.5 million) to finance their own institutional strengthening (i.e., technical assistance or studies), all other funds would go to the municipalities (para. 2.29). Project administration, monitoring and auditing (13 % of total project costs), including project execution at the municipal level and semi-annual project performance reports (para. 2.50) produced by the participating Provincial Executing Units and reviewed by the Central Coordinating Unit. D. Eligibility Criteria 2.4 In order to effectively implement the project's objectives in 13 provinces, each with unique demographic, economic, physical and political characteristics, it is essential that the eligibility criteria be clear, easily determined and fully accepted by all institutions involved in the project's implementation. These basic criteria would be defined at three levels: (a) provincial eligibility; (b) municipal eligibility; and (c) subproject eligibility. Provincial Eligibility Criteria 2.5 In order to ensure full compatibility with the objectives and conditions of the Provincial Development Project (Loan 3280-AR), provinces wishing to participate in the project would have to be in compliance with the eligibility

18 criteria of the provincial project, including the introduction of any proposed subprojects into the provinces' approved Financial Action and Investment Plan as defined in the Provincial Development Project. While all 13 provinces would be potential candidates for the project, each participating province would have to demonstrate - as an additional condition of eligibility - that it has a transparent and predictable revenue sharing (coparticipaci6n) law that transfers sufficient funds to guarantee municipal loans in accordance with procedures to be established in the Project Operations Manual (para. 2.35). An initial review of the potentially eligible provinces shows that they all have revenue-sharing laws to transfer national funds to provinces and provincial own-source revenues on to municipalities (Annex D). Although all have clearly defined percentages for primary distribution of national funds to provinces and provincial ownsource revenues to all municipalities, the secondary distribution to each municipality is not entirely transparent in two cases (Misiones and Tierra del Fuego). Furthermore, in the Province of La Rioja, the revenue-sharing law has been temporarily suspended. These conditions would have to be corrected for the three provinces to be eligible to participate in the project. 2.6 All funds onlent to municipalities would be guaranteed through their share of the provincial to municipal revenue sharing system. In addition, all loan repayments to the provinces would be channeled through a financially autonomous Municipal Development Fund (MDF). 2.7 The creation and continuation of the MDF would be a condition of eligibility for each province. The provinces would apply "harder" onlending terms to municipalities in order to capitalize the MDFs, which would then be used exclusively for further onlending to municipalities at positive real interest rates and also to provide a reserve to cover the foreign exchange risk. Each province would be autonomous in establishing its onlending policies to municipalities within the following minimum measures for MDF capitalization: (a) (b) while project loans from the Central Government to provinces would be repaid over 15 years, including 5 years of grace, provincial onlending to municipalities would have up to 10 years for loan repayment, including up to 1 year of grace; and in onlending to municipalities, provinces would include a spread of at least 1.5% over the interest rate for provincial repayment of loans to the central government to help capitalize the MDF and cover cross-currency risks. 2.8 The autonomy of the Municipal Development Funds would permit the participation of eligible municipalities in provinces lacking the creditworthiness to borrow, as such provincial governments would not be able to use these resources for any purpose other than the stated purpose of the fund (i.e., repayment of the Bank debt service, reserve to cover the cross-currency

19 exchange risk, and financing additional municipal investments). Furthermore, in such cases, the province would have to demonstrate that the fund's cash flow would be able to service in a timely fashion all the province's financial commitments arising from the project. By including municipalities in provinces lacking the creditworthiness to borrow, the project would encourage municipalities in all provinces (both creditworthy and uncreditworthy) to improve their own fiscal performance. Increased municipal resource mobilization and fiscal autonomy could reduce municipal demands for provincial transfers, which are an important and growing drain on provincial finances and impair provincial creditworthiness (para. 1.21). Municipal Eligibility Criteria 2.9 Regarding municipal eligibility criteria, all municipalities would be eligible to participate in the institutional strengthening component of the project. In order to receive credit for financing municipal infrastructure investments, however, municipalities would have to meet creditworthiness criteria, and demonstrate that the proposed investments are consistent with the municipalities' general development plans and their overall investment programs (including investments financed from other sources) With respect to creditworthiness, to be eligible a municipality would have to demonstrate that: (a) (b) (c) its current revenues (excluding discretionary grants) are greater than its current expenditures (including payments of interest and principal); and either its outstanding and proposed debt obligations are within reasonable limits, defined by the following criterion: total debt service on total outstanding and proposed loans does not exceed 15% of total budgeted revenues for the year of the proposed project; or total outstanding and proposed debt does not exceed 60% of total municipal revenues (excluding credit) of the previous year, adjusted for inflation These criteria reward municipalities in provinces with automatic revenue sharing, as municipalities in these provinces would obviously have higher current revenues (own-source plus automatic revenue sharing) and thus would be more likely to have current account equilibrium (criteria (a)) and a higher debt service capacity (criteria (b)). Given that there will be competition for loan proceeds, this may induce municipalities to pressure their provincial governments to make their provincial to municipal revenue-sharing systems more predictable and transparent.

20 Subproject Eligibility Criteria 2.12 While all municipalities would be eligible for institutional strengthening programs, to access funds for particular investments, municipalities would have to meet additional eligibility criteria that, together with the municipal creditworthy criteria (paras ), would provide adequate assurances that: (a) (b) (c) incremental resources mobilized through the project would be used in a non-deficit and non-inflationary way; municipalities that participate in the project would be financially strengthened in order to provide other services not covered by the project; and subprojects are technically, economically, financially, and environmentally sound Cost recovery is essential to all three of these criteria. A fundamental condition would therefore be that at least 65 % of the total cost of the aggregate investments proposed in a municipality in a given year be recovered from payments to the municipality by direct beneficiaries (e.g., betterment levies, connection fees, user charges, sales, concessions and rents) In addition, the Project Operations Manual would define the specific methods to be used for the technical, economic, financial, and environmental evaluation of subprojects. The evaluation methods chosen would reflect the physical risks of subprojects with different degrees of complexity and size. For example, a street lighting subproject would normally use minimum cost economic analysis, especially if there is full cost recovery. However, some larger or more complex subprojects would require full cost-benefit analysis and a complete environmental assessment. The preparation and review requirements would be defined in the Project Operations Manual. The responsibility for supervising the application of these requirements would rest primarily with the provincial executing units (paras ) Furthermore, in the case of municipalities subject to conditions of unusual physical risks (such as potential floods in the La Plata River basin), investments would only be allowed if it is confirmed that they would not be unduly threatened by such conditions To facilitate subproject evaluation procedures and promote broad institutional benefits resulting from implementing numerous, small-scale investments in many municipalities, smaller (e.g., costing at most US$2.0 million equivalent) and less complex subprojects would be given priority.

21 The project would mainly finance those physical investments that are exclusively municipal in nature. The criteria for determining this condition are: (a) (b) (c) that the investment be procured under the municipality's authority; that the municipality be responsible for its operation and maintenance; and that after the investment is completed, the municipality will have authority over service tariffs or fees Given that Argentina's decentralization program is ongoing and that the responsibilities of the municipalities continue to evolve (para. 1.10), the proposed project would make it a condition of subproject eligibility that only those subproject in sectors which are clearly the municipalities' responsibility (as defined in the Ley Orgdnica Municipal) would be eligible Force account procedures would be not be used so as to keep the project simple and to promote private sector participation in the project With respect to service-provider cooperatives which exist in some provinces for which there may be a question whether final responsibility rests with the municipality or the cooperative, investments would be considered eligible for financing under the project if: (a) all of the three criteria of the preceding 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 municipality. An exception to these conditions would be made in the case of expansions to gas distribution networks to be carried out by municipalities to extend these services to areas not covered by the gas companies. In these cases, there would have to be full cost recovery from the households benefiting from the expansions through payments attached to their regular consumption bills Although preference would be given to smaller investments (para. 2.14), there would be no limitation on the size of individual investment subprojects. The only exceptions would be that there would be a maximum allowed for investments above US$2.0 million of 30% of the loan amount, and that in the four largest provinces (Buenos Aires, C6rdoba, Santa Fe and Mendoza), the Bank would only finance subprojects up to US$2.0 million equivalent each (those above this amount would be financed by the IDB). E. Project Costs 2.22 The total Bank project cost is estimated at US$300 million equivalent. Together with the parallel IDB operation (US$300 million), the total cost of the

22 joint Bank-IDB program would amount to US$600 million. The proposed Bank loan of US$210 million equivalent would finance approximately 35% of these total costs. A summary of project costs is presented in Table 2.1. Table 2.1: Estimated Program Costs Local E2oreiz 2tl -Q uiy 1994 USS miions)- Municipal Infrastructure Institutional Strengthening Project Administration 40 = (including audits) Total Bank Project Cost Total IDB Project Cost * Total Program Cost Project cost estimates have been based on actual costs of similar items recently procured in the ongoing first Municipal Development Project (Loan 2920-AR). These costs are expressed in US dollars because of the parity between the dollar and the peso. The proposed project would finance a time slice of municipal investments. Thus, these cost estimates are only approximations, and no separate contingencies were calculated For the Bank's project, foreign exchange costs are estimated at US$90 million equivalent (30% of the total), of which approximately 89% is for municipal infrastructure and 11 % for institutional strengthening Dimensioning of project and loan sizes were based primarily on: (a) a demand survey conducted during preparation of the project; (b) the implementation capacity of municipalities in the participating provinces; (c) actual investment levels and implementation experience in the first project; and (d) the coverage by IDB of all investments above US$2.0 million equivalent in the four larger provinces financed in parallel with the Bank. Although there is an estimated potential demand for municipal infrastructure of US$1.3 billion (Annex C), it was deemed prudent to limit the amount of the project and loan in accordance with the above factors. F. Allocation of Loan Proceeds by Province 2.26 The Central Government's policy is to provide equal access to international assistance for all provinces. This policy needs to be combined with the actual potential for effective use of the available funds. Thus, the allocation of funds from this program to each province will be done according to a combination (in equal parts) of the quotas established in the distributo of revenue sharing from the Central Government to the provinces (coparticipaci6n)

23 and the distribution of population in the participating provinces. However, to maintain some flexibility and to allow for adjustments according to actual interest and use of the funds during implementation, the allocation would be done in two stages. Most of the available funds would be distributed according to the indicated formula for use during an initial period of three years. However, given that municipalities in the four largest provinces (Buenos Aires, Cordoba, Santa Fe and Mendoza) would also have access to funds from the parallel operation financed by the IDB, the initial allocation to these four provinces would be halved - the other half going into a pool of funds to be used at the second stage. Funds from the initial allocation not committed during the first three years would also go into the pool mentioned above. The allocation of funds from the pool would be on the basis of actual demand (i.e., first come, first served). Based on the above criteria, the initial allocation of loan proceeds would be as follows (in US$ thousand equivalent): Table 2.2: Initial Allocation of Loan Proceeds by Province * IBRD Loan (US$'000) = $210,000 Initial Allocation (US$'000) = $129,838 Pool of Unassigned Resources (US$'000) = $80,162 Provinces bnitial Allocation % of Initial (US$'000) Allocation Buenos Aires 46, Santa Fe 13, C6rdoba 13, Corrientes 9,621 7 Misiones 8,925 7 Formosa 7,678 6 Mendoza 6,588 5 Catamarca 5,639 4 Neuquen 4,572 4 La Rioja 4,340 3 La Pampa 4,210 3 Santa Cruz 3,275 3 Tierra del Fuego 1,417 1 * See Annex E for details. Totals 129, While some of the provinces participating in the project have had experience in the ongoing operation, many of the factors affecting implementation would still exist (e.g., time required to pass new provincial laws permitting indebtedness or to prepare new investment subprojects). Also, many of the provinces would be new to this type of program. The implementation

24 schedule of six and a half years for project completion has taken these factors into account. Annex H shows the estimated disbursement schedule. G. Financing Plan 2.28 It is proposed that the Bank finance approximately 35% of the program's total cost (i.e., including the IDB operation), with all counterpart funds in the Bank's project to be provided by the municipalities. The new IDB municipal loan to be implemented in parallel with this operation is also expected to finance 35 % of the combined costs of the program. This level of financing is justified because of (a) the need to provide strong incentives to improve the fiscal behavior of municipal governments, (b) the important and long-term impact of the institutional strengthening expected from the project, and (c) the limited capacity of municipalities to provide the necessary counterpart funds, given the strict eligibility criteria imposed on them by the project. Also, Bank financing in both the Municipal Development and the Provincial Development projects is 70%. The financing plan of the total program is presented below. Table 2.3: Financing Plan Local Foreign Tota -(uly 1994 USs milions) - Bank IDB Municipalities 180 = Total The Bank loan would be made at the Bank's standard variable interest rate with repayment in 15 years, with a grace period of 5 years. All proceeds of the Bank loan would be onlent from the Central Government to provincial governments eligible under this loan with the same terms and conditions as Bank lending to the Argentine Republic (para. 3. l(b)(i)). Provinces would assume the cross-currency exchange risk and would pay a charge of up to 2.5% of the loan amount for the operations of a central coordinating unit (para. 2.32). These conditions would be established in Subsidiary Loan Agreements to be signed between the Central Government and each participating province (para. 3.2(b)). Although provinces could use up to 5% of the amount of the loan (US$10.5 million) to finance their own institutional strengthening (i.e., technical assistance or studies), all other proceeds of the Bank loan would be onlent to municipalities. This 5 % would be allocated among the provinces in direct proportion to the initial allocation (see last column of Table 2.2) Onlending of the proceeds of the Bank loan from provinces to municipalities would be on terms and conditions satisfactory to the Bank that would be established in subloan agreements between each province and the

25 municipalities. Municipalities would provide all local counterpart funds for investment subloans Municipal repayment of loans would be guaranteed through the municipalities' funds from the provinces' revenue sharing with them (para. 3. 1(a)(ii)). H. Implementation Arrangements 2.32 Overall coordination of project execution would be the responsibility of the Under Secretariat of Housing (SV) in the Secretariat of Social Development of the Presidency (para. 3. 1(a)(iii)), but actual implementation would be the responsibility of each participating province and the respective municipalities. The SV would maintain the existing Central Coordinating Unit (CCU) for the project. The CCU would manage the Special Account, provide general coordination and support to the provincial executing agencies, manage the flow of documents required by the Bank and the provinces, monitor and report on project implementation (including the preparation of the semi-annual performance reports and the consolidation of annual financial audits from each province, paras ), and carry out several studies on generic issues affecting the performance of municipalities. The CCU would also be available to coordinate the dissemination of experience and techniques to the provinces, and to coordinate other functions on behalf of, and at the request of, the provinces. During negotiations the Government presented a preliminary plan to strengthen the CCU so it will operate with sufficient qualified staff (para. 3. l(a)(iii)). Submission of the finalized staffing plan, satisfactory to the Bank, would be a condition of loan effectiveness (para. 3.2 (e)) Each province would form a Provincial Executing Unit (PEU) that would be primarily responsible for implementation, supervision and evaluation of all project activities in the province (Annex G). The model to be followed would be the executing units established in the five provinces belonging to the ongoing Municipal Development Project (Loan 2920-AR). The PEUs would not be large entities, nor would they duplicate the functions and tasks of existing agencies. Rather, they would function under the authority of the Provincial Governors to ensure that all project activities are properly coordinated with the relevant provincial ministries and agencies necessary for successful project implementation, in addition to performing directly all planning, programming, supervision and reporting duties. The establishment and maintenance of the PEU, with functions and staffing satisfactory to the Bank, would be a condition of eligibility and disbursement in each province (paras. 3. 1(b)(ii) & 3.2(c)(i)). The operating costs of a PEU, as well as the provinces' share of the costs of the CCU and of the commitment fees, would come from regular provincial budgetary appropriations. There are two basic conditions necessary for the PEU to adequately assume its implementation responsibilities under the project: (a) adequate staffing, with sufficient technical expertise to carry out implementation

26 functions directly (para. 3.2(c)(i)); and (b) sufficient authority, ultimately flowing from the Governor, to be able to mobilize and coordinate resources across provincial agencies as well as among municipalities. These would be conditions of disbursement for each province All physical investments under the project would be fully in compliance with Bank policies on the environment, including preparation of environmental assessments and impact statements when required. As a means of ensuring compliance and consistency, the Project Operations Manual would provide environmental guidelines for evaluating subprojects (para. 3.1(b)(x)). The implementation of these guidelines will be the responsibility of the PEUs. Specifically, each PEU would be responsible for screening proposals to ensure that those which could have environmental impacts are analyzed and appropriate actions are taken. When an environmental assessment is required, the PEU would be responsible for seeing that all needed actions are taken (e.g., hiring environmental consultants to prepare the assessment and developing environmental mitigation and management plans). As a condition of eligibility, the provinces would have to demonstrate to the CCU that the respective PEUs have adequate environmental capacity to carry out the above measures Detailed implementation arrangements would be described in a standard Project Operations Manual satisfactory to the Bank, which each province would agree not to amend without the Bank's prior consent (para. 3. 1(b)(v)). In addition to the project's eligibility criteria (paras ), the Manual would include, inter alia, subloan terms and conditions, guidelines for procurement, draft tender documents, subloan disbursement procedures, guidelines and control procedures for components' unit costs, and model contracts between the province and its financial intermediary, between the province and its municipalities, as well as the main elements of the Loan Agreement and the Subsidiary Loan Agreement Municipal authorities would have to prepare and present to the PEU annual or multi-year investment plans 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 basis. Annual Investment Plans, as consolidated and proposed by the province (including provincial technical assistance and training, in addition to the municipal infrastructure and institutional strengthening subprojects), would be reviewed and approved by the Bank (para. 3. 1(b)(iv)), and only subprojects included in these Plans would be eligible for financing under the project. These Plans would need to show, for each proposed investment, cost recovery procedures and projections, as well as their individual impact on the municipality's creditworthiness and that the proposed investment is the clear responsibility of the municipality or province.. A standard format to be used by all provinces in the submission of these Plans would be defined in the Project Operations Manual in a form satisfactory to the Bank. In the case of municipalities submitting investments in high-risk areas

27 (e.g., the La Plata River flood plain), adequate assurances that these risks have been considered would need to be presented as well As a condition of effectiveness, at least one province would have to sign a Subsidiary Loan Agreements with the Central Government (para. 3.2(b)) A Supervision Plan is provided in Annex G, and includes a Project Launch Workshop, bi-annual supervision missions, and annual visits to Washington by CCU staff. This Annex also includes a matrix showing the key indicators for the implementation and development objectives that would be monitored during supervision missions. I. Procurement 2.39 Based on the experience of the first municipal development project, problems with procurement matters would be minimal given the small size of most subprojects and the use of standard bidding documents satisfactory to the Bank provided in the Project Operations Manual. To facilitate procurement, PEUs would be required to hire competent procurement review consultants to assist them in implementing procurement efficiently. Procurement for civil works and goods costing US$350,000 and US$100,000 equivalent or less, respectively, would be done through international or local shopping as applicable. This is expected to account for approximately US$25 million equivalent (US$18 million for civil works and US$7 million for goods), or 8% of total project costs. Civil works costing between US$350,000 and US$5,000,000 equivalent and goods costing US$100,000 up to US$350,000 accounting for US$184 million (US$143 million for civil works and US$41 million for goods), or 61 % of total project costs, would be carried out through I,ocal Competitive Bidding (LCB) procedures acceptable to the Bank. Procurement of all civil works and goods costing US$5,000,000 and US$350,000 equivalent or more, respectively, accounting for US$25 million (US$18 million for civil works and US$7 million for goods), or 8% of total project costs, would be through International Competitive Bidding (ICB) in agreement with the Bank's Guidelines for Procurement under IBRD Loans and IDA Credits. Consultant services, estimated to cost US$34.3 million equivalent, or 11 % of total project costs, would be procured in accordance with the Bank's Guidelines for the Use of Consultants by World Bank Borrowers. The use of LCB procedures and standard bidding documents for goods and works acceptable to the Bank, and of standard documents for the selection and hiring of consultants, would be a condition of the loan. These procedures would be in accordance with all Bank policies (para. 3. 1(b)(vi)). In ICB procurement, as well as the contracting of consulting services, the Bank's standard documents would be used. These arrangements are summarized in Table 2.4 below and would be included in the Project Operations Manual for each province.

28 The provinces participating in the proposed project would exempt contracts for goods, works and services to be financed under the project from all legal and regulatory provisions which are in conflict with the Legal Agreement's procurement provisions for goods, works and services (para. 3.2(c)(ii)). Table 2.4: Procurement Method by Category (in US$ million equivalent) ICB LCB Other NBF Total Civil Works (18) (107) (13) -- (138) Goods Consulting Services (7) (31) (5) --- (43) (a) Included in eligible subprojects under project (22) (22) components (a) and (b) (b) For the CCU including travel and per diem (5.2) -- (5.2) Auditing Services (1.5) --- (1.5) Training _- (0.3) -- (0.3) Project Administration (0) (0) 1DB Project (0) (0) Total of which IBRD financed (25) (138) (47) (0) (210) * Numbers in parentheses are the anounts financed by IBRD. * Other includes S25 mnillion for local and international sbopping and employment of consultants in accordance with World Bank guidelines. * NBF = not Bank financed PEUs would review, ex-ante, all procurement procedures, documents, bid evaluations, and contract awards related to civil works and goods to ensure that the agreed procurement processes are 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 reviews would be delegated, in the first instance, to each province's PEU. However, in order to ensure compliance with Bank procurement policies, each PEU would submit for ex-ante review by the Bank, all documentation to be used for the procurement

29 of civil works contracts costing US$1.5 million or more, including bid documents, evaluations and recommended contract awards. Similarly, each PEU would submit for ex-ante review by the Bank, all documentation to be used for the procurement of goods contracts costing US$150,000 or more. Each PEU would review ex-ante all procedures and documentation to be used for contracting consultants. The Bank would review ex-ante all consultant contracting procedures and documentation for all contracts costing US$100,000 equivalent or more in the case of firms, and US$50,000 or more for individuals, as well as the terms of reference of all consulting assignments. Table 2.5: Procurement Prior Review and SOE Thresholds (in USS '000 equivalent) Prior Review SOE Civil Works All contracts 21,500 Contracts < 1,500 Goods All contracts 2150 Contracts <150 Consulting Services Firms 2100 Firms <100 Individuals 250 Individuals < The performance review reports required for each province (para. 2.51) would provide certification that the procurement presented to the Bank on a sample basis for ex-ante review is typical and representative. Furthermore, these reports would provide certification that all other procurement was also carried out in accordance with the agreed procedures. In all cases, however, the Bank would review, on an ex-post basis by sampling, LCB, international and local shopping, and contracting of consultants procurement procedures. If it were determined that procurement was not made following the agreed procedures, then no expenditures for such items would be financed out of the proceeds of the loan, and the Bank may cancel the corresponding loan amount. The financial audits would also include separate opinions on the compliance (in each province) with procurement and on compliance with cost recovery obligations (para. 2.50). J. Disbursement 2.43 For withdrawal of proceeds of the proposed Bank loan, the following categories and disbursement procedures are proposed: (a) for eligible civil works, 100% of foreign expenditures and 75% of local expenditures; (b) for equipment, materials and vehicles, 100% of foreign expenditures and 75% of local expenditures; and

30 (c) for consulting services, including necessary travel, 100% of foreign and local expenditures For municipal expenditures in categories (a), (b) and (c) above, disbursements would be authorized only after verification and certification by the PEU (according to procedures established in the Project Operations Manual) that expenditures resulted from a previously approved subloan agreement between the municipality and the province, and that the expenditures correspond to a subproject included in a previously approved Annual Investment Plan (para. 2.36) Because of the large number of contracts and relatively small payments, disbursements for certain local expenditures would be made against Statement of Expenditures (SOEs). All civil works below US$1.5 million and all goods contracted below US$150,000 would be disbursed using SOE procedures. Contracts with consulting firms and individuals estimated to cost less than US$100,000 and US$50,000 each, respectively, would also be disbursed against SOEs. Disbursements from the Loan Account for contracts above the thresholds detailed above would take place only after the Bank gives its no objection to the procurement procedures used SOEs would contain expenditures of subprojects with a breakdown of outlays for equipment, materials, consultants' services and civil works under contract. Documentation of these expenditures would not be submitted to the Bank but would be retained by the PEUs for periodic inspection by Bank staff. Disbursements for other expenditures would be made against full documentation. Each PEU would maintain adequate records for municipal expenditures, and each would prepare consolidated SOEs for periodic transmittal to the CCU and subsequent submission to the Bank In order to avoid the dismantling of the already functioning PEUs in the five original provinces, and to encourage provincial initiative for rapid and effective institutional support for project implementation, retroactive financing not to exceed US$21.0 million in total is recommended for expenditures incurred after March 1, 1994 and not earlier than 12 months before loan signing in those provinces that comply in substance with all the terms and conditions to be defined in the Loan Agreement - particularly those related to procurement procedures The estimated disbursement schedule is based on historical data and experience gathered under the first project, and the characteristics of the participating provinces and municipalities. The schedule assumes some expenditures eligible for retroactive financing beginning March 1, 1994 and loan effectiveness in June This schedule is presented in Annex H. Accordingly, the project's completion date would be December 31, 2001, and the closing date would be June 30, 2002.

31 A Special Account, in US dollars would be established in the Banco de la Naci6n Argentina and managed by the CCU (para. 3. 1(a)(i)). The authorized Special Account allocation would be US$15 million. It was agreed during negotiations that the amount of the initial allocation would be US$5 million. K. Accounts and Audits 2.50 Separate Project Accounts in a financial intermediary would be opened and maintained in each participating province and municipality. This obligation would to be established in the corresponding Subsidiary Loan Agreements and Subloan Agreements. The Special Account (para. 2.49) and the Project Accounts in each province and municipality would be audited by a single independent auditor acceptable to the Bank (paras. 3.1(a)(iv) and 3.1(b)(vii)). For municipalities and PEUs, an annual audit of the financial statements of the Project Accounts, including a separate opinion on withdrawal applications made on the basis of SOEs, would be furnished to the Bank not later than six months after the end of each fiscal year. These reports would also include separate opinions about the degree of compliance with procurement procedures and cost recovery obligations assumed by municipalities when submitting investment proposals for approval, as well as separate opinions on compliance with the obligations related to the Municipal Development Funds. In addition, each province would provide to the Bank through the CCU semi-annual performance review reports. Terms of reference, satisfactory to the Bank, for conducting the performance review reports would be a condition of effectiveness (para. 3.2(d)). These reports would provide evaluations of the compliance by all participating provinces and municipalities with all legal agreements - including the provisions of the Project Operations Manual - reached with the Bank, province, provincial intermediaries, and municipalities (para. 3.1 (b) (viii)). The CCU would review the provincial reports and provide its own comments to the Bank. The performance review reports from each participating province, would be delivered to the Bank not later than March 31 and September 30 of each year. The CCU would send the Bank its independent opinion on the contents of the provincial reports, by April 30 and October 31 of each year The semi-annual performance reviews are essential to project design in order to make effective Bank supervision manageable in the face of a large number of subprojects spread over an enormous geographic area with numerous political and administrative jurisdictions. Because of the demanding audit and review requirements of the project, and because of the need to ensure timely compliance with all legal agreements and to minimize risks of interruption of disbursements through SOE procedures, the Bank would finance 100% of the performance review and audit expenditures. If the financial or performance review identifies significant irregularities or lack of compliance with any of the obligations mentioned above in any municipality, then the Project Operations Manual would provide that, as a condition of receiving project funding for municipal investments in the subsequent audit period, the province would

32 provide a program (including an institutional strengthening program if appropriate), satisfactory to the Bank, focused on eliminating the observed irregularities or issues. If these problems persist for another audit or review period, the municipality would be declared ineligible for subsequent project financing (paras. 3.1(b)(ix)). L. Flnancial Impact 2.52 Municipal creditworthiness and component eligibility criteria would ensure that all municipalities that participate in the project would be financially strengthened because of their participation. For those municipalities that will be eligible initially only for institutional strengthening assistance - because they do not meet the creditworthiness criteria - the project would focus provincial assistance on improving municipal finance performance so that they may be eligible for physical investment loans in the future From the perspective of direct project beneficiaries, the project's financial impact would be positive if two key assumptions are met: (a) affected families and firms are rational and accurate in estimating that proposed infrastructure benefits, as capitalized in their property values, will be greater than the costs they must pay (generally through betterment levies) in order to receive the service; (b) beneficiaries agree to finance infrastructure improvements on a voluntary basis. Because the experience with past financing of municipal infrastructure in Argentina indicates that both of these assumptions are reasonable, the risk of negative project financial impact on direct beneficiaries is very low The only provincial expenditures that would be required by the project are counterpart funds for the provincial institutional strengthening program, operation of the PEU, contribution for operation of the CCU, and commitment fees. Even if the provinces spend 5% of the total amount of the loan for their own needs, this amount would be less than 0. I% of the annual revenue of any of the provinces. As these costs would be affordable for the provinces, project implementation delays resulting from insufficient provincial counterpart funding would not pose a significant risk. M. Economic Impact 2.55 The weighted average of the internal economic rate of return (IERR) for a sample of paving, street lighting, and sanitary subprojects is 19.7% (Annex F). The three sectors evaluated represent 54% of the total estimated potential demand of all sectors. The average IERR for each of the subsectors are: 17.8 % for paving; 21.1% for sanitation; and 28.8% for street lighting. These three sectors represent 31%, 21% and 2% of total potential demand, respectively.

33 N. Poverty Impact 2.56 While not explicitly designed to do so, the project would mostly benefit the urban poor, who would be the primary recipients of the municipal infrastructure financed by the project. Specifically, the project would have a significant social impact through: (a) financing an increase in "adequate" service levels for water and sanitation; (b) providing increased affordability for these and other municipal services (such as road paving and street lighting) because of the project's relatively long-term credit financing; and (c) generating employment for low-skilled labor during construction By providing access to municipal infrastructure, the project would also remove constraints on the potential productivity of the urban poor in terms of physical access (road paving), personal safety (street lighting), and health and environmental hazards (sanitation and refuse collection). 0. Environmental Impact 2.58 The project has been defined as category B for environmental purposes, since some physical investments could have specific environmental impacts. The net environmental impact, however, is expected to be positive. The project would fund the means for evaluating a subproject's environmental impact in each participating province as noted in para Further, through street paving the dust from unpaved streets (a major contributor to air pollution in many areas) would be reduced. The project would also bring about, through sanitation subprojects, a reduction in soil and water contamination, as well as improved erosion control through drainage subprojects. P. Risks and Safeguards 2.59 Risks in this operation would be limited because of the experience gained in the first project and the similarity between these two operations. The main identified risk is the possibility that some provinces may prove ineffective or not continue their interest in the project. To safeguard against this risk, the project would be spread among 13 provinces, five of which (Buenos Aires, C6rdoba, La Pampa, Neuquen and Santa Fe) participated in the first operation and were the main promoters of the new project. Thus, should there be some provinces not interested or capable of fully participating in the project, the large demand for municipal infrastructure and the flexibility built into the design of the procedure for allocating funds would permit shifting loan funds to the provinces with more capacity and interest. No special risks have been identified with regard to financial or technical matters, as most of the investments to be financed - as in the first project - would be small and simple and limited to eligible municipalities. A minor risk is the inclusion in the project of municipalities in areas affected by possible floods. To reduce this risk,

34 municipalities would have to provide, as a prior condition, certifications from the corresponding PEUs that the proposed investments would be in safe areas and were designed in accordance with appropriate standards. (A study is being completed for the Subunidad de Control Contra Emergencias of the Ministry of Interior that will provide guidelines for these areas).

35 m. AGREEMENTS REACHED AND RECOMMENDATION A. Assurances 3.1 During negotiations, assurances were obtained as follows. (a) From the Central Govermment that: (i) (ii) it will open and maintain, in US dollars, a Special Account in the Banco de la Naci6n Argentina (para. 2.49); all of the proceeds of the Bank loan will be onlent to eligible provinces in accordance with an allocation formula acceptable to the Bank (para. 2.26) under terms and conditions equivalent to the Bank loan, with each participating province's revenue sharing serving as a guarantee for repayment of subloans (para. 2.31); (iii) it will maintain a Central Coordinating Unit within the Under Secretariat of Housing of the Secretariat of Social Development of the Presidency, and that this entity will provide overall project implementation supervision, including independent reviews of the provincial semi-annual performance reports and preparation of semi-annual progress reports on the overall project (para. 2.32); and (iv) it will have the Special Account and the Project Accounts in each province and municipality audited according to standard country-wide procedures satisfactory to the Bank (para. 2.50). (b) That the Central Government will cause the provincial governments to: (i) (ii) (iii) assume their share of the loan on the same terms and conditions as received by the Central Government and onlend to municipalities on terms and conditions satisfactory to the Bank (paras. 2.7, 2.29 and 2.30); establish, adequately staff and otherwise support, in a manner satisfactory to the Bank, Project Execution Units during the entire project execution period (para. 2.33); finance the operations of the Central Coordinating Unit (para. 2.32);

36 (iv) (v) (vi) agree to submit timely Annual Investment Plans (including municipal investments and provincial institutional strengthening) for Bank review and approval (para. 2.36); approve and supervise the program carried out by the municipalities in accordance with the Project Operations Manual, and not amend this Manual without the Bank's prior consent (para. 2.35); carry out procurement in accordance with Bank policies (paras ), and provide separate annual certifications to this effect in the performance review reports (para. 2.42); (vii) have the financial statements of the project account in each province and participating municipalities, along with cost recovery measures, audited annually by a single auditor acceptable to the Bank (paras and 2.51); (viii) have semi-annual evaluations of performance, including the compliance with all agreements entered into under the project, provided through performance review reports prepared on the basis of terms of reference acceptable to the Bank (paras. 2.42, 2.50 and 2.51); (ix) if significant deficiencies in implementation are identified in audits of the provincial project accounts and/or the PEU, appropriate institutional strengthening assistance for the PEU would be proposed as an obligatory part of the subsequent year's investment program (paras and 2.51), and if significant irregularities in municipal accounts and/or performance are identified in the periodic audits (inter alia, cost recovery), further project funding for physical investments in that municipality would be made conditional on an institutional strengthening program, satisfactory to the Bank, focused on eliminating the observed irregularities; and (x) comply with all necessary environmental protection requirements, as provided in the Project Operations Manual (para. 2.34). B. Conditions of Effectiveness 3.2 The following would be conditions of effectiveness: (a) Bank approval of the Project Operations Manual (para. 2.35);

37 (b) (c) the execution of a Subsidiary Loan Agreement between the Central Government and at least one participating province (para. 2.37); that the province referred to in (b) above has: (i) (ii) (iii) established and adequately staffed the Provincial Executing Unit (PEU) (para. 2.33); 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 (para. 2.40); and presented a description of specific environmental procedures and institutional responsibilities that is consistent with the guidelines in the Project Operations Manual and acceptable to the Bank. (d) (e) that terms of reference, satisfactory to the Bank, for conducting the performance review reports (including a review of performance indicators) would be presented (para. 2.50). that a staffing plan, satisfactory to the Bank, would be presented for the CCU to ensure that it is sufficiently strengthened to perform its assigned functions (para. 2.32). C. Condition of Disbursement 3.3 A condition of disbursement for expenditures to be made by any province not included in 3.2(b) would be that it also meet the conditions in 3.2(b) and (c) above. D. Recommendation 3.4 With the above assurances and conditions, the proposed project is suitable for a Bank loan of US$210 million equivalent, to be repaid over a period of 15 years, including a five-year grace period, at the Bank's standard variable interest rate.

38

39 ARGENTINA SECOND MUNICIPAL DEVELOPMENT PROJECT STAFF APPRAISAL REPORT ANNEXES A. Previous Bank Experience in the Sector B. Population Growth and Distribution C. Demand for Municipal Infrastructure D. Intergovernmental Fiscal Relations E. Initial Allocation and Commitment Fees by Province F. Economic Analysis..58 G. Implementation and Supervision Arrangements.. 61 H. Disbursement Schedule.. 65 I. Selected Documents and Data Available in Project File.. 66

40

41 Annex A Previous Bank Experience in the Sector 1. The proposed project is similar to the newer style of municipal development projects that stresses the improvement of efficiency and fiscal discipline of the municipal public sector, such as in the first Argentina (Loan 2920-AR) and Ecuador (Loan 3285-EC) operations. As these municipal-finance oriented operations are relatively new, there are only a few Project Completion Reports (PCRs) and Project Performance Audit Reports (PPARs) available on this type of operation. Also, a recent report issued by OED (7Wenty Years of Lendingfor Urban Development, ) concentrates its analysis primarily on urban issues as defined in the previous traditional type of urban projects, referring primarily to the relationship between these projects and the key items now identified by the Bank for its urban policy agenda: improvements in the urban environment, poverty reduction, economic growth and increased efficiency of infrastructure services. However, the report also comments that "from the outcomes of completed operations there is evidence of impacts through real sector, fiscal, and financial links...but there is scope for improvements in 'operationalizing' these links." The proposed new project has been designed with these comments in mind. This annex will review Bank reports on such completed projects, and then comment on ongoing projects of this type in the Department, taking into consideration the different risks that will need to be analyzed. Because the proposed project has little in common with the older, multi-sector urban projects, these projects will not be covered in this review. 2. In reviewing the experience of past projects and the lessons learned in order to apply them to the proposed project, it is useful to distinguish among different types of potential risks, depending on the project's objectives. Municipal Development projects usually include: (a) policy reform objectives; (b) investment objectives; and (c) institutional strengthening objectives. The first two are often interconnected by using investments as incentives to induce policy reforms, while institutional strengthening components are used to facilitate achievement of the other two. The focus of the Bank's analysis has often concentrated on risks related to the investment objectives, primarily because of the particular importance of disbursements and because the associated targets are relatively easy to quantify and monitor. However, particularly in projects designed to be responsive to fiscal matters, it is also important to address risks associated with the other two types of objectives and of the corresponding components. Lessons from Completed Projects 3. The newer project style evolved as the Bank moved away from 'multisector investments in one or a handful of urban areas to a more narrow, multi-

42 Annex A city, subsectoral approach" (OED, Annual Review. 1990, p. 37). These projects also reflect a new policy agenda 'bringing together the concepts of economic growth, efficiency, environmental impact, and poverty reduction" and charting a "more comprehensive approach that relies much more on institutional development and reform at the municipal and national levels" (OED, Twenty Years of Lending for Urban Development, : 1994, p. vii). Under the first of these studies, OED reviewed four such projects in Korea, Ecuador, and Brazil. These projects focused on institutional development but still placed a heavy emphasis on physical investment, usually in smaller-scale urban infrastructure. According to this report, this "newer-style subsector type of urban project proved to be particularly satisfactory." 4. Of these projects, the experience of the Parana Market Towns Project (PRAM) is particularly relevant to the proposed new project. The PPAR for this project concludes: Municipal access to funds for a wide variety of urban investments stimulated community participation and reportedly made local authorities more accountable to their constituents. Availability of long-term credit permitted municipal governments to finance projects whose costs could be recovered, thus helping to enhance local revenues. While it is difficult to assess project impact on local economic development, the operation clearly helped to improve the quality of life in small towns and cities throughout Parana. PRAM was also successful from a financial standpoint as there were no reported cases of subloan default or significant arrears. PRAM's success can be attributed to a number of factors: (a) it was based on an existing local initiative that was able to maintain strong political support even when control of the state government shifted from one party to another; (b) most subprojects were small and of relatively simple design and easy to implement; (c) it was administered by a competent and motivated team and was able to utilize the decentralized structure of the existing state municipal assistance agency; (d) it was able to meet the priority demands of a large number of municipalities, providing it with additional political support; (e) as a multi-town program, risks of delayed or non-execution were spread over a large number of sub-borrowers; and (f) its location in a state having a tradition of good public administration provided a propitious institutional setting for project execution (Report No , paras ). 5. The later OED study (Tventy Years of Lending) was a much more extensive study, reviewing 115 completed urban projects in 52 countries. Many of its conclusions repeat and confirm earlier observations. Specifically, it recommends the following lines of action: (a) tapping the potential of real,

43 Annex A fiscal, and financial linkages of urban projects to enhance their impact; (b) strengthening the congruence of objectives and design; (c) securing project ownership (by the beneficiaries); and (d) improving monitoring and evaluation systems. 6. A recent PCR for a municipal development project in Honduras (Loan 2583-HO) concludes with a number of useful lessons: (a) the need for continuing close coordination between donor agencies; (b) the need to address the system of incentives within which local governments operate; (c) institutional strengthening programs must be provided within an appropriate environment and sufficient time horizon; (d) the desirability of incorporating private capital markets to mobilize and allocate savings for municipalities; (e) having well designed decentralization policies; and (f) the need for close Bank supervision in the case of programs supporting institutional reform. Lessons from the First Municipal Development Project in Argentina (Loan 2920-AR) 7. Many of the above comments from the OED report are also applicable to the first Municipal Development project in Argentina, currently under implementation. While the project suffered serious delays in the beginning, primarily because of the macroeconomic difficulties at the time ( ), it quickly recovered and is now scheduled to be completed within the original time period (completion date June 30, 1995). In terms of its investment objectives, it has been extremely successful in the timely provision of urgently needed urban infrastructure. Its main achievement in terms of its policy objectives was the introduction of simple, but effective fiscal conditions as preconditions for access to investment funds. In so doing, the project will have made a significant contribution to the fiscal objectives of the Government. Indeed, all subloans have been fully repaid on time by the municipalities, maldng it unnecessary for the provinces to draw on municipal coparticipaci6n funds. Lastly, with respect to institutional strengthening, this was not a primary objective of the project and the corresponding component has been only partly successful. Nevertheless, the project has been able to establish and maintain strong and effective provincial units with good administrative, managerial, and operational capabilities. This has been a major factor in the successful implementation of the project. It has also begun a process of strengthening these capabilities in the participating municipalities. 8. Reflecting the experience and lessons of other municipal development projects, the success of the Argentina project in terms of investments has been due primarily to the following factors: (a) the concentration on small, simple and easy to implement investments; (b) the coverage of a large number of interested municipalities; (c) the maintenance of very competent local executing

44 Annex A units; and (d) the flexibility of shifting and reallocating loan funds among competing provinces. 9. A related factor in the success of this project has been the quite limited role assigned to the federal level. In fact, there is only a one-person coordinating office in charge mainly of routine contacts with the Bank and some very limited review functions. This has avoided the delays usually created by a large, controlling central unit, and greatly contributed to the efficient implementation of this project. 10. Many of the factors listed above were also crucial to the achieved policy objectives. In addition, this project benefitted by having only a few and simple "pass - no pass" fiscal conditions. These were easy to quantify, easy for the municipalities to understand, and simple for the Bank to review and approve. 11. In terms of institutional strengthening programs, the limited success in this project can be attributed primarily to the initial lack of understanding (by both the provinces and municipalities) of the link between such programs and the ability of municipalities to expand their borrowing capacity. This was because these programs were promoted and justified on their own merits, rather than as a means to facilitate the achievement of other project objectives and goals. The lack of interest in this component was also affected by the reluctance of municipalities to borrow for these types of programs. However, the provinces have carried out many institutional strengthening programs with their own resources, and therefore, the limited use of loan funds may not properly reflect the actual extent of these activities. Strategy 12. Given the above experiences and lessons, the proposed project would follow, with respect to investment objectives, a strategy containing these elements: (a) (b) (c) maintain the principle of competition among provinces and municipalities to enable loan funds to be reallocated away from "non-performing" to more agile units; keep most individual investments simple, easy to prepare, justify, review and procure; establish simple and standardized procurement procedures and documents;

45 Annex A (d) (e) (f) (g) promote the use of larger consulting firms for standard procurement processes; complete the preparation of investments for at least the first year of implementation prior to negotiations; delegate all possible implementation responsibilities to the individual provinces, while maintaining only the smalluest possible central unit; and ensure the existence of proper incentives for participation in the project, particularly by insisting on sufficiently transparent and predictable revenue sharing systems as a condition of eligibility, and by limiting discretionary funds (i.e., those without conditions) available to municipalities. 13. With respect to policy objectives, the proposed project would try to integrate the following factors: (a) (b) (c) limit conditions related to policy reforms to only a few selected and reasonable items; define these reforms and conditions in simple terms, preferably of the "pass - no pass type (i.e., which do not require extensive studies, analysis, or complicated formulas); and require only conditions which will be significant but not so onerous as to encourage municipalities to seek discretionary funds from other sources. 14. With respect to institutional strengthening activities, the proposed project would: (a) (b) (c) emphasize the importance of these programs as a vehicle for improved capacity to obtain investment funds; promote the use of standard assistance packages coordinated (and perhaps prepared and financed) at the provincial level; and seek ways to reduce the cost of these activities, e.g., through matching grants from the provinces or through bilateral grants.

46 Demographic Profile Annex B Population Growth and Distribution 1. Argentina is one of the most urbanized countries in Latin America, with the unique characteristic that urbanization was well advanced even before World War II. In 1991, 87% of the of the total population lived in urban areas of 2,000 or more inhabitants. Despite its high level of urbanization, Argentina is still experiencing significant growth in its urban population, which grew from 23.2 million in 1980 to 28.4 million in 1991 (1.85% per year) as shown in Table B. 1. Conversely, rural population decreased at a rate of % per year over the same period. 2. Argentina is also one of the Latin American countries in which the urban population is more heavily concentrated, with only three cities (Buenos Aires, C6rdoba and Santa Fe) comprising over 41 % of the total population. This concentration of the population in a few major urban areas contributes to the high density disparities among provinces, ranging from 41 inhabitants per km 2 in Buenos Aires to 0.7 inhabitants per km 2 in Santa Cruz (see Table B.2). 3. The 13 provinces participating in the project constitute 70% of the total population. According to their size, these provinces can be divided into two major groups: (a) four highly-populated provinces (Buenos Aires, C6rdoba, Mendoza, and Santa Fe) with 19.6 million inhabitants (60% of the total Argentine population); and (b) nine less-populated provinces (Catamarca, Corrientes, Formosa, La Pampa, La Rioja, Misiones, Neuquen, Santa Cruz, and Tierra del Fuego) with 3.3 million inhabitants (only 10% of the total population). Distribution of Municipalities 4. In Argentina, municipalities are recognized by the National Constitution as an independent level of government. However, the criteria that determine municipalities are defined by each province in its Carta Orgdnica. Thus, there are as many different definitions of municipalities as there are provinces. 5. Overall, municipalities can be distinguished into two groups according to their geographic base. (a) Territorial municipalities. These municipalities cover a geographical territory. In each territorial municipality there can be one or more urban areas. Buenos Aires, Mendoza, and La Rioja have adopted this type of municipalities.

47 Annex B (b) Urban municipalities. These municipalities are urban areas, and their geographic limits coincide with those of the urban area. In the provinces which have adopted this type of municipality, the provinces have jurisdiction over rural areas. Under this definition, city size is generally the criteria used to define a municipality, with the particular population threshold varying greatly from province to province. Ten of the provinces participating in the project have urban municipalities. 6. Table B.3 shows the total population and number of potential participants in the program by province. All urban areas with 500 inhabitants or more have been included in the case of provinces with urban municipalities, as each province can potentially make all urban areas eligible to participate in the project, independently of the particular population criteria specified in the Carta Orgdnca.' Including both territorial and urban municipalities, there are a total of 912 potential participants in the program, with more than 21.7 million inhabitants. 1 Tis was the case in Santa Fe under the first Municipal Development Project.

48 Annex B Tble B.1: Urban & Rural Population by Province, _.... "'.'. ' j ) J j N10..., f. -.,... j..,,.0,-0-...,--.. Totali 23,191 28, % 4,737 4, % IBRD/IDB Provinces Buenos Aires 10,123 11, % % C6rdoba 1,944 2, % % Mendoza 824 1, % % Santa Fe % % Subtotal 14,913 17, % 1,865 1, % IBRD Provinces Catamarca % % Corrientes % % Formosa % % La PaMpa % % La Rioja % % Misiones % % Neuqu6n % % Santa Cnuz % l.55% Tieffa del Fuego _ % % Subtotal 1,551 2, % 1,369 1, % Other Provinces 3,804 5, % 1,503 1, % Fedeal Capita 2,923 3, % % Soume: INDEC C.nus of Populatio 1980 and 1991.

49 Annex B Table B.2: Population, Density & Growth by Province I ; E ',: Popua6 t-, - - :, si.:.er. ' Provi 9 19 Growth Ra (% g q. hm.) s. 991) Totals 27,948 32, ,758, IBRD/IDB Provinces Buenos Aires 10,865 12, , C6rdoba 2,408 2, , Mendoza 1,196 1, , Santa Fe Subtotal 16,935 19, , IBRD Provinces Catamarca , Corrientes , Formosa , La Pampa , La Rioja , Misiones , Neuquen , Santa Cruz , Tierra del Fuego Subtotal 2,514 3, , Other Provinces 5,576 6, ,138, Federal Capital 2,923 2, ,827.0 Source: INDEC Census of Population 1980 and 1991.

50 Annex B Table B.3: Number and Population of Potential Participants by Province 'Provinces NOber:aot Potential Population 0 Participants: (i fin0ot 1991J TOTAL ,659.3 Buenos Aires ' ,582.3 Catamarca Cordoba 219 2,475.0 Corrientes Formosa La Pampa La Rioja L' Mendoza!Y 18 1,414.1 Misiones Neuqudn Santa Cruz Santa Fe 235 2,516.3 Tierra del Fuego / Buenos Airms, La Rioja, and Mendoza have territorial municipalitics. Thus, there mfuay be one or more urban areas within one municipality. Sources: INDEC Census of Population 1991 and data from provinces.

51 -43- Annex C Demand for Municipal Infrastructure Unmet Needs for Municipal Infrastructure 1. As shown in Table C. 1, urban agglomerations in the participating provinces still have considerable unmet needs for municipal infrastructure. There is greater need for paving and sanitation systems than for water and street lighting. For instance, the percentage of those without paved roads ranges from 68% in Misiones to 26% in Bahfa Blanca. Similarly, the percentage of those without sewerage mains ranges from over 75 % in C6rdoba and Misiones to 18 % in Santa Cruz. 2. In terms of the spatial distribution of unmet infrastructure needs, there is a higher absolute need in the three largest provinces (Buenos Aires, Santa Fe, and C6rdoba), which have over half of the total number of poor households in the country. Although the percentage of unmet infrastructure needs is higher in provinces with smaller populations, there are fewer households than in the largest provinces. Potential Demand for Municipal Infrastructure 3. The estimated potential demand was based on a demand survey conducted during preparation of the project. The survey was administered in 70 municipalities in the participating provinces, which were asked among other items to identify infrastructure needs or subprojects. These subprojects, in turn, were assigned a cost following a well-defined methodology based on existing data from similar subprojects. Only highly regarded subprojects were included in the sample. Once the sample was defined, it was expanded to estimate total demand. The data collected from larger municipalities, mainly those in the larger provinces, was not expanded because the sample covered all of these municipalities. 4. The estimated potential demand for basic municipal infrastructure in the 13 provinces is approximately US$1.3 billion, of which 90% are subprojects under US$1.5 million. The municipalities' identified needs can be classified in 17 sectors, ranging from water connections to cemeteries. However, most of the subprojects identified are in the street paving and sanitation sectors, which account for 31% (US$406 million) and 21% (US$281 million) respectively. In the four provinces in which both banks (IBRD and IDB) would work in parallel, subprojects with an estimated cost of US$2.0 million or more equivalent were removed from the sample, since these subprojects would be executed by the IDB. The estimated demand includes US$90 million equivalent in subprojects that were identified during the first Municipal Development Project, but because of lack of funds, could not be implemented.

52 Annex C Table C.1: Infrastructure Needs in Major Urban Centers and Metropolitan Areas by Province (1988) Provinces Urban Center Of liowehol4& wivth Seiri'e Outsidc the or Metropolitan Area.,, fpavin Waler Sewag e t Average vstret Maim ZIainm Lighting Buenos Aires Gran Bahia Blanca Gran La Plata Catamarca Gran Catamarca C6rdoba Gran C6rdoba Corrientes Corrientes n.a Formosa Formosa La Pampa Sta. Rosa y Toay La Rioja La Rioja Mendoza Gran Mendoza Misiones Posadas Neuquen Neuquen Santa Cruz Rio Gallegos Santa Fe Gran Rosario Sta. Fe y Sto. Tomd Tierra del Fuego Usuahia y Rfo Grande n.a. n.a. n.a. n.a. n.a. Average Source: Siruaci6n y Evolaci6n Social, Sintesis. NMmero 2, INDEC, 1993.

53 Municipal Flnancial Indicators Annex D Intergovernmental Fiscal Relations 1. Based on 1992 municipal finance data from a stratified sample of 44 municipalities in the 13 Bank provinces representing over 20% of the total urban population in those provinces, the following indicators were estimated for each municipality. (a) (b) (c) Current Account Equilibrium: This was estimated as Current Revenues (excluding discretionary transfers) / Current Expenditures. Table D. 1 shows the percentage of municipalities which fall within each range. Overall, current revenues exceed current expenditures in 75 % of the municipalities. However, current expenditures exceed current revenues by up to 25 % in the remaining one-quarter of the municipalities. By this criteria, 75 % of municipalities would be eligible for investments under the project. Total Stock of Debt / Current Revenues (excluding discretionary transfers): This ratio which reflects the degree of indebtedness of municipalities shows that overall, municipalities hold very little debt. Less than 7% of all municipalities hold a total stock of debt which exceeds 60% of their current revenues. In other words, 93% of the municipalities would have sufficient debt capacity to participate in this project, using this ratio. Debt Service (Interest + Amortization) / Current Revenues (excluding discretionary transfers): This indicator shows that current debt service does not represent a significant burden for the large majority of municipalities. Less than 3% of all municipalities pay more than 15 % of their current revenues to service current debts. Thus, 97% of municipalities would have sufficient debt capacity to participate in the project, using this indicator. Provincial to Municipal Revenue Sharing 2. Although provincial to municipal revenue sharing relationships are similar to those between the central and provincial governments, each province determines the particular allocation criteria and percentages used for both primary and secondary distribution in a coparticipaci6n law. Table D.2 shows the share of the revenues from provincial taxes and federal coparticipaci6n that is automatically transferred to the municipalities as a whole, i.e., the primary

54 Annex D distribution. For the 13 project provinces, the primary provincial to municipal distribution of total national automatic revenue sharing ranges from 5% in Corrientes to 30% in Tierra del Fuego. For provincial taxes, the percentages going to municipalities ranges from 14% in Buenos Aires and Mendoza to 809% in Neuqu6n. In some provinces, such as Catamarca, La Rioja, Santa Fe the percentages vary for different provincial taxes, while in Santa Cruz property and vehicle taxes are under municipal jurisdiction. 3. The criteria used in the secondary distribution in provincial to municipal revenue sharing can be classified as redistributional and as fiscal incentives. In the case of redistribution criteria, they can be characterized as proportional or as compensatory. Proportional redistribution criteria are those positively correlated with the municipality's tax receipts, such as population, land surface, and equal share. Conversely, compensatory redistribution criteria are those inversely correlated with the municipalities' tax receipts, such as the inverse of population and basic needs; thus, they have a greater redistributive impact. In the case of fiscal incentives, the allocation of shared funds is based on a measure associated with the "tax effort" of the municipality, such as tax revenues and increases in tax receipts. Table D.3 shows that all provinces use proportional distribution criteria, while Corrientes, La Rioja, Mendoza and Santa Cruz also benefit less populated areas. Corrientes is the only province that does not apply fiscal incentives as an allocation criterion. 4. Table D.4 shows the percentage of automatic national revenue sharing funds that are transferred to individual municipalities based on distributive and fiscal criteria. They vary greatly, from 0% in Corrientes to 64% in La Pampa. 5. Based on data from the same 44 municipalities described above, the following indicators were estimated for each municipality: (a) (b) Automatic Revenue Sharing Transfers (coparticipaci6n) / Current Revenues (excluding discretionary transfers): This indicator reflects the proportion of municipal revenues automatically transferred to the municipalities which are relatively predictable and based on percentages established by coparticipaci6n laws effective in each province. Overall, Table D.5 shows that almost 39% of municipalities receive between 40% and 60% from these sources. Discretionary Transfers (both grants and loans) / Total Revenues (excluding discretionary transfers): This indicator reflects the dependency of individual provinces on discretionary funds, which are fluctuating and quite unpredictable. These transfers represent less than 5% of current revenues for the majority of

55 Annex D municipalities (64%). However, some provinces such as La Rioja have a chronic dependence on this type of source. Municipal Debt Capacity 6. Based on data from the same 44 municipalities, debt capacity as shown in Table D.6 was estimated for each municipality as 60% of Current Revenues minus Total Stock of Debt. Debt capacity was subsequently added by province to obtain debt capacity for the sample at the provincial level. This was then divided by the urban population represented in the sample to obtain the per capita urban debt capacity by province. The total municipal debt capacity in each province was estimated by multiplying the corresponding per capita urban debt capacity by the total urban population in the province.' The overall debt capacity is almost US$2.5 billion, with municipalities in the four largest provinces (Buenos Aires, C6rdoba, Mendoza and Santa Fe) holding almost 95 % of the total debt capacity. I dc = (.60 * CR) - TD where dc is smple Debt Capacity, CR are Current Revenues and TD is Total Debt; dc(p) = SUM {dc) where dc(p) is the overall sample municipal Debt Capacity by province; DC, = DC(P) / up where DC, is the per capita urban debt capacity by province, UP is the urban population in the sample; DC(P) = DP, * UP where DC(P) is the overall debt capacity by province, UP is the urban population in the province.

56 Annex D Table D.1: Percentage of Municlpalies By Current Account EquilIbrhum (1992) Curfent Account Equilibrium 100.0% (Current Revenues Y' / Current Expenditures) 80%-90% 15.9% 90%-95% 6.8% 95%-100% 2.3% 100%-110% 18.2% +110% 56.8% Total Stock of Debt / Curret RevenuekW 100.0% 0-2% 61.4% 2%-S% 9.1% 5%-10% 4.6% 10%-20% 9.1% 20%-30% 6.8% 30%-60% 2.3% +60% 6.8% Debt Sevice 2' / Current Revenues k' 100.0% 0-5% 90.9% 5%-10% 6.8% 10%-15% 0.0% 15%-20% 0.0% +20% 2.3% Basd on a umpb of 44 municipalities in the province comrpooding to do proqct. hi excludes discrtionary tansfe. / kit includes amortization. Source: Data cooected from province by Dank mission.

57 Annex D Table D.2: Criteria for Provincial to Municipal Revenue Sharing of National Automatic Revenue Sharing Funds and Provincial Taxes: Primary Distribution Buenos Aires Catanuarca 10.0" 8.5 C6rdoba Correntes ~' Formosa LA Pampa La Rioja ' 10.0 Mendoza 4' Misiones Neuqu& 4 ' Santa Cruz LI Sat Fe Tierra del Fuego 4' el V Y di In Cataffarca, municipalities receive 10%, 28% and 70% from revenues from property, gross income, and vehicle taxes. T'hewe percentages determine minimumn level of funding. In LA Rioja, municipalities receive 20 % and 50 % of revenues from gross income and vehicle taxes respectively. Mendoza, Neuqu6n, Santa Cruz and Tierra del Fuego also receive a percentage of oil revenues. In Santa Fe, munjicipalities receive % of sale tax revenues and national automiatic revenue sharing fuends, as wel a 50.0% and 90.0% of revenues fr4om property and vecle taxes respectively. Source: Lay Nacionai , 1928

58 -50 - Annex D Table D.3: Criteria for Provincial to Municipal Revenue Sharing of National Automatic Revenue Sharing Funds and Provincial Taxes: Secondary Distribution -: ::: :: sxt : : i : ::JR Buenos Aires x x x x Catamarc-a x x x C6rdoba x x x Corrientes x x Formosa x x x La Pampa x x x La Rioja x x x Mendoza x xx Misiones L' Neuqu&n x x x Santa Cruz x x x 2 SantaiFe x x x Tieffa del Fuego x x x jj I/ Information on the criteria uaed in secondary distribution in Mfisiones is noa svailable. Misiones' Ley de Coparicipaddn determines scondary distribution based on fixed coefficients basd on criteria not explicitly specified. In Santa Cniz, couection of property an vehicle taxes is under municipal jurisdiction. Source: Ley Nacional , 1988

59 Annex D Table D.4: Percentage of National Sharing Funds to Municipalities Allocated Based on Hscal Effort: Secondary Distribution -rb~ie % Automai : Crtia Used in Sodary Diddl *a -: -Na-i:n: Re,eoe Redstbuto-- - -'~ : Buenos Aires Catamarca C6rdoba Coffientes Formosa La Pampa La Rioja Mendoza MisionesW n.a. n.a. Neuqu6n Santa Cruz Santa Fe Tierra del Fuegof n.a. n.a. / smisioner' Ly de Cop wicfpad&s determines econdary distribution based on fixed coefficients not explicitly specilied. Data on percentages for the allocation criteria uwed for secondary distribution in 1arra del Puego ae misring. Sources: Provincial Laws gaverning revenue sharing.

60 Annex D Table D.5: Percentage of Municipalities By National Automatic Revenue Sharing and Discretionary Transfers as a Proportion of Current Revenues (1992) Automatic Revenue Sharing / Current Revenues 100.0% 0-20% 15.9% 20%-40% 34.1% 40%-60% 38.6% 60%-80% 6.8% +80% 4.6% Discretionary Transfers / Current Revenuesa 100.0% 0-5% 63.6% 5%-10% 9.1% 10%-20% 11.4% 20%-40% 6.8% 40%-60% 2.3% 60%-80% 6.8% a/ Based on a sample of 44 municipalities in the provinces corresponding to the project. b/ It excludes discretionary transfers. Source: Data collected from provinces by Bank mission.

61 Annex D Table D.6: Debt Capacity for All Municipalities (in 1992 USS),...''.',""':,,,.,,....,,., T,.,',,.,. t.,.,.,, :1 :-W.-,.b. ;, N.. ~~~~~~~~~~~~~~~~~..... TOTAL 4,267, ,177,000 2,960.2 Bhueos Aires 1,259, ,894,000 2,098.6 Catewulca 114, , C6rdoba 1,363, ,332, Corientes 361, , Formosa n.a. n.s. 281,000 n.a. LA PaMpa 10, , La Rioja 118, , Mendoza 547, ,100, Misiones 286, , Neuqudn 7, , Santa Cuz n.a. n.a. 147,000 n.a. Santa Fe 150, ,391, Tioa del Fuego 32, , It Based on a mnple of 44 municipalitie in the provinces coneqonding to the proe. Debt capacity calculated as (.60 RC) - D} where RC are curret account fids, uad D is currat debt. Sourc: Data collected fiom provinc by Bnk minion.

62 Initial Allocations Annex E Initial Allocation and Commitment Fees by Province 1. About US$129 million of loan funds will be allocated in the first stage lasting three years. This represents the allocation of funds to each province based on a formula which gives equal weight to: (a) the provinces's participation in the revenue sharing system of the Central Government; and (b) the province's population size. The only exception would be that the largest four provinces (Buenos Aires, C6rdoba, Santa Fe and Mendoza) would receive only half of their theoretical allocation because they would be receiving an equal amount from the companion IDB loan. The remaining amounts would be part of the unallocated resource "pool" (see Table E. 1). The reason for having an initial assignment is to allow all provinces a fair chance to use the loan proceeds. If there were no initial assignments, the more efficient provinces might use all of the resources before the less efficient ones could act. Less efficient provinces might not even try to compete. The resources in the "pool" would be available to any province that had committed all of its initial assignment to approved subprojects on a first come, first served basis. 2. After the three-year period, all the initially allocated resources not committed to approved subprojects will also go to the "pool" to be allocated on a first come, first serve basis. The "pool" then rewards the provinces that are more efficient in subproject execution and introduces an element of competition into the assignment of resources among the provinces. Provinces have to utilize their initial assignments of loan proceeds by the end of the three-year period or compete with other provinces for them once they become part of the "pool." 3. The following criteria were used in allocating the half of the loan that would initially be assign among the provinces. (a) Coefficients of Coparticipaci6n: Half of the initially assigned amount is allocated among the potentially eligible provinces based on the secondary distribution coefficients used in national to provincial coparticipaci6n.' These coefficients are strongly redistributive, favoring the poorer and less populated provinces at the expense of the more developed and heavily populated ones. 1 The coefficient used for each province is obtained by dividing the coefficient of coparticipacion of the province by the sum of the coefficients of coparnicipacion of all the participating provinces.

63 Commitment Fees Annex E (b) Population: the other half is allocated among the potentially eligible provinces based on the distribution of population - a proxy for actual potential demand. 4. Provinces are expected to pay an 0.25 % commitment fee for those funds that are initially allocated and have not yet been disbursed. Table E. 1 shows what provinces would pay in commitment fees based on the initial loan allocation and assuming that no disbursements were made after a year. Should initially allocated funds that were not used by a province transfer to the general pool after three years, the province which makes use of the funds will be responsible for past commitment fees. All provinces will pay commitment fees on the resources in the "pool" in accordance with the same criteria used in the initial allocation with compensation by those provinces that actually use the resources. Ilitial Loan Allocation, Potential Demand, and Debt Capacity 5. Table E.2 shows the comparison between initial loan allocation, potential demand and debt capacity for all participating provinces. From the comparison, it is clear that potential demand and debt capacity exceed initial loan allocations by an ample margin. Total estimated potential demand exceeds initial loan allocation by eight times for the province of Santa Cruz, and by as much as 35 times for the province of La Rioja. Similarly, debt capacity exceeds initial loan allocation by 1.5 in the case of La Rioja, and up to 50 times in the case of Tierra del Fuego. These results indicate that the funds made available through this loan would be used by the participating provinces.

64 Annex E Table E.1: Initial Assignment and Commitment Fees Based on National to Provincial Revenue Sharing (50%) and Population (50%) Distributions IBRD Loan = $210,000,000 Initial Assignment = $129,838,243 'Pool' of Unassigned Resources = $80,161,757 Iff Commitment Fee = 0.25% 1: SP,,as tf ( WAJ%f *I t of [A+13ji2 ID olrd al C m- -- A C 1ip f -f; -10 ; Aver sahualsla do 4Uca io)v p,,hihd.'' Totas % % % $129,838,243 $324,500 22,918 $14.83 IBRD/IDB Provinces Buenos Aires 33.62% 54.96% 44.29% S46,504,189 S116,260 12, Santa Fe 13.68% 12.21% 12.95% S13,594,558 S33,986 2,798 $4.36 C6rdoba 13.60% 12.07% 12.83% S13,475,359 S33,688 2,767 S4.87 Mendoza 6.38% 6.16% 6.27% S S $4.66 Subtotal 67.29% 85.40% 76.34% $80,161,757 S200,404 19,573 $4.10 EEBRD Provinces Corrientes 5.69% 3.47% 4.58% $9,621,400 S24, $12.09 Misiones 5.06% 3.44% 4.25% $8,925,030 $22, $11.31 Fonnosa 5.57% 1.74% 3.66% $7,677,790 $19, $19.27 Catamarca 4.22% 1.15% 2.69% $5,638,597 $14, $21.34 Neuqu6n 2.66% 1.70% 2.18% $4,571,678 $11, $11.76 L.a Rioja 3.17% 0.96% 2.07% $4,340, , $19.66 La Pampa 2.88% 1.13% 2.00% $4,210, , $16.19 Santa Cruz 2.42% 0.70% 1.56% $3,274,828 $8, $20.49 Tierra del Fuego 1.05% 0.30% 0.67% S S $20.42 Subtotal 32.71% 14.60% 23.66% $49,676, ,191 3,346 $14.S5 a/ IDB will finane half of the initial allocation of the four largest provinces; hence the anmounta shown are one half of thd tot amol t available to the four under the initial allocation of the joint progam. Sources: CoparicipacOn Distribution from Dirccci6n Nacional de Coordinacin Fiscal con a Provvnd as, and Populati(o data from INDEC Census of Population, 1991

65 Annex E Table E.2: Total Potential Municipal Demand for Loan Proceeds, Municipal Debt Capacity, and Initial Loan Allocation by Province (in 1992 US$ millions) I Tot Debt I tia Lan PFWToalsi 1 Danand W Capacity l' Assignnte s t TOTAL 1, , Buenos Aires , Catamarca C6rdoba Corrientes Formosa 81.0 n.a. 7.7 La Pampa La Rioja Mendoza Misiones Neuquen Santa Cruz 13.5 n.a. 3.3 Santa Fe Tierra del Fuego a/ Total etimated demand is calculated bcaed on a stratified sauple of 107 municipa ities in the participating provinces and adjusted for total population by province (Annex C). b/ Debt capacity is calculated based on financial data from a tratified sample of 44 municipalities in the participating provinces and adjusted by total population by province (Annex C). c/ Initial loan allocation among provinces is based on two different criteria: (1) 50% is calculated based on the coefficients used to asign national revenue-sharing funds among provinces; and (ih) 50% is baeed on total population in the provinces.

66 Annex F Economic Analysis 1. The economic evaluation of the program was done based on a sample of 17 sanitary, paving, and street lighting subprojects representative of about 54% of the total potential demand.' The overall weighted internal economic rate of return (IERR) shown on Table F. 1 was estimated in three stages as follows: (a) IERR by subproject: The IERR was estimated using one of the following benefit-estimation methods: Hedonic Prices: In the hedonic price method, marketed commodities such as housing are considered to have a range of attributes or services, with its implicit prices being reflected in the total commodity price paid. The method derives estimates of these implicit prices by observing how the price of the commodity varies with its attributes. Using the hedonic price method, the implicit value of a particular urban service, such as street paving, street lighting, or sewage connections, is derived by observing how property values vary depending on whether the service exists or not. Contingent Valuation: The contingent valuation method is useful to derive the value of non-marketed commodities for which relevant behavior is not observable. The method overcomes the lack of an observable market by presenting respondents with an hypothetical market. Using contingent valuation, the value of a particular service improvement is derived based on respondents' stated choices in an hypothetical market that presents them with the improvement in question, the institutional structure under which it would be provided, etc. Their responses are taken as valid indicators of how these individuals would behave if such market existed. Expected Losses: In the expected losses method, the value of a particular subproject is derived based on all the losses (i.e, property damage, foregone income, etc.) that would be avoided as a result of such subproject over its life expectancy. Expected losses are estimated as the value of the loss multiplied by the probability of its occurrence. For instance, in the case of a Potential demand was estimated based on a stratified sample of 107 municipalities in the participating provinces and adjusted by total population by province (see Annex C).

67 Annex F flood-control project that is expected to last 50 years, the value of the subproject is estimated as the value of all the losses that would result from flooding over the 50-year life of the project, affected by the probability of their occurrence (i.e., the losses that would result from a 100-year flood would be valued at half their value). (b) Weighted [ERR by subprojects in the subsector: For each of the three subsectors represented in the sample (i.e., sanitation and flooding, urban roads, and lighting works), the average IERR was estimated by weighting the average IERR of each subproject by its cost relatively to the total cost of all subprojects in the subsector. In particular, IERR,P() = E ierr(,, * c(,, / C(,> where IERR(,) is the weighted rate of return for each subsector in the sample, ierr(;) is the subproject rate of return, c<;> is the cost of the subproject, and C(,) is the sum of the costs of all subprojects in the subsector, namely, C(,) = E c(d. (c) Overall IERR weighted by subsectors in the project: This was estimated by weighting the average IERR of each subsector in the sample by the cost of the subsector in the project relatively to the total cost of the project. In particular, IERR< = E ierr( 8 ) * c(s) / C(p) where IERRQ, is the weighted rate of return for the overall project, ierr( 5 ) is the subsector rate of return based on the sample of subprojects, c(s,) is the cost of the subsector in the project, and Cp) is total costs of the project.

68 Table F.1: Internal Economic Rates of Return for Sample Projects Average Rate of Retumn Weighted by Investment Cost = 19.7% Estimated Total Potential Demand as % of Sectors Covered = 53.5% Estimated Total Potential Demand = $1,326.4 million intenul A. R of R Sgtore 0% inveatmest Econosk Weg*,ad by of Total No. of Cost Rat of la.w Cot of PotastT Type of SubprqJect Decsriptiou Province Municipalty Bene&larie (US$'O00) Retum Smple Daund Totals 340,650 $53, % S3.S% I Sanitation/Flood Pump station and dike Entre Riot C. de Uruguay 12,500 $3, % 2 Sanitation/Drainage Collector and Canals in the North Sector Catamarca Catanarca 9,900 S % 3 Sanitation/Drainage Collector and Canals for the city Chubut Pto. Madryn 17,500 $5, % 4 Sanitation/Drainage 2nd, 3rd & 4th stages, South sector of city C6rdoba Rio Cuarto 55,000 $9, % 5 Sanitation/Drainage Network of collectors for the city C6rdoba Villa del Rosario 9,500 $2, % 6 Sanitation/Drainage Network of couectors for the city Entre Rios Concordia 26,500 $2, % 7 Sanitation/Drainage Network of collectors for the city Corrientes Paso de Los Libres 4,300 $ % Subtotal 135,200 $24, % 21.1% 8 Urban Roads Paving, drainage and a vehicle bridge C6rdoba BelIville 27,000 $2, % 9 Urban Roads Paving and illumination C6rdoba Marcos Juarcz 4,000 $2, % 10 UrbanRoads Pavingof96blocks EntreRios Crespo 3,200 $3,1S3 14.1% 11 UebnRoads PavinganddrainageofIllblocks EntreRios Nogoya 21,500 $5, % 12 Urban Roads Paving of 70 blocrk San Luis San LuiS 110,000 52, % 13 UrbanRoads Paving of 70 blocks and ilurnination Mendoza Tupungato 3,500 S1, % 14 Urban Roads Paving and drainage of 100 blocks Santa Fe Constituci6n 9,500 $4, % 15 Urban Roads Paving and drainage of 106 blocks Rio Negm Viedma S % Subtotal 191,300 $26, S% 30.6% 16 Lighting Mumination improvemewa for 200 blocks Rio Negro Atlen 7,700 $ % 17 Lighting IlluminAtion improvements for 230 blocks San Juan San Juan $ S% Subtotal 14,150 $2, % 1.8% Source: Project Preparmtion Team of the Sccretary of Social Development.

69 Annex G Implementation and Supervision Arrangements 1. Overall supervision responsibility for the project would rest with the Borrower's Ministry of Economy, but actual implementation would be the responsibility of each participating province and the respective municipalities. Overall coordination of project execution would be delegated to the Under Secretariat of Housing (SV) in the Secretariat of Social Development of the Presidency. The SV has established a Central Coordinating Unit (CCU) for the project and the parallel IDB-supported operation. It will provide general coordination and support to the provincial executing agencies, managing the flow of documents required by the Bank and the provinces administering the Special Account and all actions related to disbursements and audit reports, as well as monitoring and reporting on project implementation (including independent opinions on the semi-annual performance reports prepared by each province). The CCU would also coordinate the dissemination of experience and techniques to the provinces, carry out a number of key studies related to the financial and managerial situation and needs of municipalities, and coordinate other functions on behalf of, and at the request of, the provinces. The specific functions of the CCU have been detailed in the Project Operations Manual, and a corresponding document is available in the Project File. 2. Each province would form a Provincial Executing Unit (PEU) that would be primarily responsible for implementation, supervision and evaluation of all project activities in the province. The PEUs would not be large entities, nor would they duplicate the functions and tasks of existing agencies. Rather, they would function under the authority of the Provincial Governors to ensure, in addition to performing directly all planning, programming, supervision and reporting duties, that all project activities are properly coordinated with the relevant provincial ministries and agencies necessary for successful project implementation. The PEUs would also furnish semi-annual reports, commenting on the performance by the Province and its Municipalities with regard to their obligations under the respective subsidiary Loan Agreement and Subloan Agreements, including compliance with procurement procedures. The model to be followed would be similar to the executing units established in the five provinces belonging to the ongoing Municipal Development Project (Loan AR). A detailed description of these functions has been inserted in the Project Operations Manual and is available in the Project File. 3. Because each province has its own constitution, and administrative organization, and because the real structure of authority may be different from the formal one, there is no ideal institutional model for the PEU. There are, however, common institutional needs for which specific arrangements will be made in each province. The most important of these needs are as follows:

70 Annex G (a) (b) Project implementation requires inputs from many sectors. Sectoral management varies in each province, but typical examples of project implementation inputs would be: verification of proper use of provincial design standards for municipal physical works proposals; ex-ante review of all procurement procedures and certification of proper execution of civil works; verification of municipal creditworthiness; design of municipal training programs to improve financial management; and design of municipal technical assistance programs to improve maintenance procedures. The PEU would have a staff adequate to the task of coordinating intersectoral subprojects and would have ready access to specialized agencies (e.g., the Ministry of Public Works). For this access, it is necessary for the PEU to act directly or indirectly with the authority of the provincial Governor (generally the only provincial official with broad intersectoral authority). In addition, for the PEU to have effective access to the sectoral authorities along with the municipalities, it must have mobility, in both the physical and the financial sense (i.e., to visit the municipalities as well as the financial resources to react quickly to implementation needs without undue dependence on other institutions). 4. There are, therefore, two basic conditions necessary for the PEU to adequately assume its implementation responsibilities under the project: (a) adequate staffing, with sufficient technical expertise to carry out implementation functions directly; and (b) sufficient authority, ultimately flowing from the Governor, to be able to mobilize and coordinate resources across provincial agencies as well as among municipalities. 5. Supervision of project implementation would be monitored through semiannual missions to Argentina complemented by semi-annual progress reports and by periodic visits to Washington by key officials from the CCU. Because of the project's dispersed nature, the Bank would budget about 40 staff-weeks (including locauly recruited staff or consultants) to provide adequate oversight of project performance. A "project launch workshop" would coincide with the first major Bank supervision mission. The tentative Supervision Plan is shown in Table G. 1.

71 Annex G Table G.1: Supervision Plan tes Earticipant- Main: Objectivesfas May 1995 Task Manger Review of Project Operations Manual and Operations Specialist other conditions of loan effectiveness. Sep/Oct 1995 Visit by CCU staff Review of project start-up and actions to be taken by initial provinces participating in the project. Nov/Dec 1995 Task Manager Launch workshop and review of eligible Operations Specialist provinces and subproject pipeline. Review of Procurement Specialist subproject implementation pipeline, as well as Disbursement Officer implementation of procurement procedures, accounting procedures, and Progress Reports Mar/Apr 1996 Task Manager Operations Specialist Detailed review of program in a few provinces and municipalities Jul/Aug 1996 Visit by CCU staff Review progress and difficulties to date; define required actions Sep/Oct 1996 Task Manager Mid-term Review Financial Analyst Others as required Mar/Apr 1997 Task Manager Municipal Engineer Others as required Supervision tasks as required Jul/Aug 1997 Visit by CCU staff Detailed review of implementation Sep/Oct 1997 * Task Manager Supervision tasks as required Financial Analyst Others as required * Addition missions would follow the same pattern until six months before project completion, at which time an Implementation Completion Report mission would be undertaken.