Report No. 21557-AR Argentina An Assessment of Institutional Capacity for Social Sector Reform in Argentina September 17, 2001 Human Development Department Argentina, Chile, Paraguay and Uruguay Country Management Unit Latin America and the Caribbean Regional Office Document of the World Bank GOVERNMENT FISCAL YEAR January 1 - December 31 CURRENCY EQUIVALENTS Currency Unit = Argentine Peso US$1.00 = AR$1.00 ACRONYMS AND ABBREVIATIONS ANSSAL Administraci6n Nacional de Seguros de la Salud/National Health Insurance System GDP Gross Domestic Product GoA Government of Argentina LAC Latin-America and the Caribbean PRESSAL Programa de Reforma del Sector Salud Provincial/Provincial Health Sector Development Project SDP School Development Project Vice President: David De Ferranti Country Director: Myrna L. Alexander Sector Manager: Ana-Maria Arriagada Country Sector Leader: Ariel Fiszbein Task Team Leader: Donald Winkler ARGENTINA AN ASSESSMENT OF INSTITUTIONAL CAPACITY FOR SOCIAL SECTOR REFORM TABLE OF CONTENTS Executive Summary ............................................................i 1. Chapter, Background and Methodology for the Study ........................................................1 A. Introduction ...........................................................1 B. Rationale for the Study ............................................................2 C. Issues to be Addressed .......................................................... 3 D. Methodology of the Study ...........................................................5 E. Roadmap of the Study ...........................................................9 2. Chapter, Change and Reform in the Social Sectors .......................................................... 10 A. Determinants of Change .......................................................... 10 B. Institutional Context ...........................................................11 C. Provincial Context ........................................................... 13 D. Social Sector Reforms in Argentina .......................................................... 23 3. Chapter, Education: Provincial Experience ........................................................... 26 A. Educational Reform .......................................................... 26 B. Implementation of the Reform .......................................................... 30 C. Provincial Capacity ........................................................... 39 D. Capacity of the School .......................................................... 42 4. Chapter, Health: Provincial Experience in Implementing a Pilot Reform ......................... 46 A. Reform of Hospital Management .......................................................... 46 B. Implementation of the Reform .......................................................... 50 C. Provincial Capacity .......................................................... 55 5. Chapter, Lessons Learned ........................................................... 59 A. Summary of Findings .......................................................... 59 B. Lessons Learned from Education and Health ........................................................... 63 C. Conclusion .......................................................... 65 Bibliography .......................................................... 67 List of Fi2ures: 1.1 Relation Between Government Effectiveness and GDP Per Capita, All Countries 2.1 Relation between Control of Corruption and GDP per capita (PPP), All Countries 2.2 Relation between Voice and Accountability and GDP per Capita (PPP), All Countries 2.3 Actual and Predicted Government Effectiveness in Argentina 2.4 How Successful is the Ministry in Fulfilling its Mission Objectives? 2.5 Consistent Policies and Internal Communication and Ministry Effectiveness 2.6 Open and Merit-Based Recruitment and Fulfilling the Ministry's Mission Successfully 2.7 Performance Incentives and Fulfilling the Ministry's Mission Successfully 2.8 Clear Performance Standards and Fulfilling the Ministry's Mission Successfully 2.9 The Potential Effects of Incremental but Uniform Improvements in all Aspects of Capacity 3.1 Years of Educational Attainment and GDP per Capita 3.2 Third Grade Language Achievement and GDP Per Capita, Latin America 3.3 Relation between Seventh Grade Language Achievement Test Scores and Education Expenditure per Student, 1998 3.4 Relation between Seventh Grade Mathematics Achievement Test Scores and Education Expenditure per Student, 1998 3.5 Percent Changes Between Population and Secondary School Enrollments, 1994-1998 3.6 Average Language Achievement Test Scores, Grade 7, 1994-1998 3.7 The Province's Evaluation Results Strongly Affect: 3.8 Federal Education Policy: One Size Fits All 3.9 Education Transfers as a Share of Education Expenditures 3.10 Financial Consequences of Implementing the Reform 3.11 Teachers Dominate the Preparation of the School Development Project 3.12 The School Has Little Say in Teacher Training 4.1 Relation Between Population NBI and Infant Mortality Rate 4.2 Relation Between Infant Mortality Rate and Public Health Expenditure Per Capita List of Tables 1.1 Characteristics of the Provinces 2.1 Several Indicators 2.2 Provincial Revenues and Debt 2.3 Political Indicators for the Provinces 2.4 Determinants of Perceived Effectiveness and Their Impacts 2.5 Characteristics of Health and Education Reforms 3.1 Indicators of Provincial Implementation of Education Reform 3.2 Stakeholders Views of Reform Elements 3.3 The View of Provincial Education Ministry Officials: Federal Education Responsibilities are Limited 4.1 Correlation Between Per Capita Household Income and Years of Life Lost per 1,000 Population in Argentina 4.2 Quality of Maternal Care by Household Income per Capita 4.3 Health Insurance Coverage, Public Health Expenditures, and Private Health Expenditures by Income Quintile 4.4 The Characteristics of Public Hospitals by Degree of Autonomy 4.5 Hospital Autonomy in the Provinces, 1998-2000 AN ASSESSMENT OF INSTITUIONAL CAPACITY FOR SOCIAL SECTOR REFORM IN ARGENTINA EXECUTIVE SUMMARY Introduction. Argentina is a federal republic with twenty-three provinces plus the city of Buenos Aires. Beginning in the 1970s, the country began decentralizing the finance and delivery of education, health, and social welfare programs to the provinces. In the 1990s, the Menem administration initiated a number of far-reaching reforms, including reforms of the social sector and provided assistance to the provincial governments responsible for implementing them. As shown in the report, these reforms have had some successes but so far have not achieved fully all their goals. This study examines the relationship between institutional capacity at the provincial level and the reforms as one of the main factors in explaining what the constraints are and in making recommendations to continue to advance in reforming social sector services in Argentina.. To do that, we look at: (a) the economic, political, and institutional context within which the reforms were implemented; (b) the design or content of the reforms; (c) the implementation strategy of the Federal Government; (d) and the capacity of provincial governments to manage the change. Three provinces-Catamarca, Cordoba, and Salta-are studied in depth., Information on the views of stakeholders and assessments of implementation capacity were collected through interviews, focus groups, and questionnaires administered to provincial ministry officials. Additional information was provided by province-level public expenditure reviews done as part of project preparations. Drawing on prior research on the implementation of reforn, capacity is assessed by attempting to answer six questions: (a) Do leaders offer a clear vision for the future? (b) Are the ministry officials responsible for the day to day implementation of reform committed to it? (c) Does the province have the needed tools and authority to take action? (d) Are there adequate and appropriate incentives? (e) Do mechanisms exist to ensure accountability? (f) Are there adequate and predictable resources? The Context of Reform. Reforms are implemented in a particular economic, political, and institutional context, which was different in the three provinces studied. All three provinces had fiscal deficits and high indebtedness. Although Catamarca had the most serious problem, Cordoba and Salta acted first to adopt policies to reduce their deficits. Politically, Salta's government was aligned with the national administration, while Cordoba was the most strongly opposed In terms of overall provincial capacity to manage and implement policies and programs, civil service staff in all three provinces believe corruption in government is a serious problem in Argentina; that political ties are important for career advancement; that i ministries seldom have clear and consistent goals; that there are few rewards for exemplary performance or penalties for bad performance; that the performance standards required for clear accountability are generally lacking; and that financial resources are unpredictable. There are some regional differences, but, overall, provincial capacity is weak. The Reforms. The education and health reforms studied here differ significantly in their formulation and design. The health reform considered here is the introduction of hospital autonomy in management of personnel and budget -only one partial aspect of what would be considered a comprehensive reform of the health system. The education reform considered is the implementation of a comprehensive change in the education system. In terms of formulation, hospital autonomy was the result of an executive decree, and had little visibility and relatively low support among health professionals. The education reform, in contrast, was the result of lengthy debate resulting in a Federal law and high visibility. In terms of design, provinces were mandated to implement several elements in the education reform, while hospital reforms were voluntary. In terms of Federal implementation strategy, the education ministry was more proactive than the health ministry in supporting reform, although both reforms received international finance and technical assistance. Stakeholders. Important stakeholders can either support or resist reform efforts. Governors can play an especially important role in providing active leadership, most evident in the selection of ministers and setting of budget priorities. Salta's Governor initiated administrative reforms and advocated health reforms more far-reaching than those supported by the national government. Unions, also, wield significant power in Argentina's provinces, and the provincial government strategies for dealing with fiscal deficits have influenced their positions. Salta won the support of unions for reform efforts by eliminating salary arrears and guaranteeing employment (in the case of contracting out health services), while Cordoba lost union support through policies that reduced compensation. The workers-teachers, physicians, nurses, and others-who deliver social services were not strong reform supporters in general, in part because they fear the consequences. And, in the case of health, some unions ended up supporting hospital autonomy because they saw the potential for higher wages. The potential beneficiaries of reforms-patients and parents-were poorly informed and took little interest in reform, perhaps because they have almost no voice. While the education reform advocated increased participation in principle, the involvement of parents and parent associations was not necessarily welcomed by other stakeholders. Education Reform. Following years of national debate and an important Pedagogic Congress, the Argentine Congress passed legislation reforming education in 1993. While the responsibility for implementing the reform lay primarily with the provinces, the Federal Ministry of Education took a decidedly proactive role in guiding and assisting implementation efforts. This included seeking financial support from both the Inter-American Development Bank and the World Bank. The study looks at the background leading up to the reform, assesses the Federal implementation strategy, analyzes provincial capacity to manage education and ii educational change, reviews the role and capacity of the school to implement internal changes, and asks what might be the design and implementation strategy of a second generation reform. The education reform of 1993 called for expanding compulsory schooling, raising quality through curriculum reform and teacher training, involving parents in school development projects, and improving equity. The Federal Government's implementation strategy included using the Federal Education Council comprised of provincial ministers to develop policy consensus; providing financial assistance (Social Education Plan and the Federal Pact) to the provinces and schools; offering technical assistance to provincial ministries (PREGASE), and communicating information, including school outcomes, to the schools. Provincial officials, however, view much of the Federal assistance as being supply driven and failing to address provincial needs. The provinces in general complied with legal mandates, increased enrollments, and increased provincial spending on education. However, there is little evidence that the reform has had an impact on teaching and learning within the school. School development projects have not been financed adequately; teacher training has been supply-driven and had little effect; and parental participation has not materialized Provincial government and school capacity to implement the refonr is limited. There are few incentives for either provincial or school officials to implement reform. In general, information flows one direction-from the top down. While school inspectors can communicate problems upwards as well as facilitate program implementation, there are few of them, and they often lack adequate transportation. While the provinces have significant authority to make educational decisions, the schools have very little formal authority, and school directors have limited levers to bring about change. A second stage reform to improve teaching and raise quality will require more complex design and implementation. The overall weak capacity of the average provincial education ministry could seriously limit the success of a second stage reform. Hospital Reform. In one of a series of executive decrees aimed at improving efficiency in the health care system, the Federal Government in 1993 issued Decreto 578/93 which reaffirmed the right of the provinces to give public hospitals management autonomy and allowed for the automatic cost recovery by public hospitals from health insurers through the national social security fund. Subsequently, the Government created a project, with World Bank support, to provide technical assistance to provinces to implement the decree in selected public hospitals. This study looks at the background leading up to the reform, assesses Federal and provincial roles in implementing the reform, presents some of the results, and ends with a discussion of how this pilot reform might be taken to scale. The Federal strategy to implement hospital reforms relied on technical assistance (PRESSAL) combined with a "carrot" of infrastructure improvements for selected provinces making progress, although in the end it became too difficult to tie capital investment to hospital autonomy. The Federal Government did not effectively iii communicate the content of the reform and its rationale to stakeholders, resulting in opposition based on a fear of the unknown. Provincial health ministries were not strongly committed to hospital reform, which threatened to take important decisions concerning appointment of hospital directors, procurement, etc., out of the hands of the ministry. An important exception is Salta, where the leadership of the Governor ensured a minister and ministry strongly committed to reform. However, even where commitment was strong, other measures of provincial ministry capacity were weak. If hospital reform were to be taken to scale, a more strategic implementation plan would be required on the part of the Federal Government. The Federal minister could use the Federal Health Council (COFESA) of provincial health ministers as a vehicle to bring about commitment and permit learning between the provinces. An information campaign targeted on provincial ministries and physicians and health workers would also be necessary. Lessons Learned. This study covers a small sample-two reforms in three provinces-which argues for caution in generalizing about lessons learned. On the other hand, these lessons confirm implementation experiences reported earlier in other countries. One lesson is to confirm an old one, that leadership counts, especially when that leadership is provided by the governor of a province. A second lesson is that it's difficult to obtain commitment from actors (e.g., teachers, physicians) who have had no participation in reform design. A third lesson is that for Federal reforms to succeed where Federal finance is small and provides little leverage requires a sound communication and information strategy that builds a broad-based demand for reform, or at least diminishes the fear of change. A fourth lesson is to not impose the costs of reform on those who are required to implement it. A fifth lesson is that Federally- initiated reforms are difficult to implement if the Federal ministry itself does not take a proactive stance. A sixth lesson is that Federal reforms that are top-down in their implementation face a formidable obstacle in the low effectiveness of most provincial ministries. A final lesson is that the support or opposition of stakeholders can be strongly influenced by the implementation strategy chosen by the province. The largest dilemma facing both Federal and provincial governments in implementing reform is how to improve the capacities of provincial ministries. The high degree of political influence and perceived corruption; the lack of relationship between merit and career advancement; the lack of horizontal and bottom-up communication are problems that will require a long time to resolve, and the implementation of reform cannot wait for their solution. Given this weak provincial capacity, there are strong arguments for Federal strategies to provide assistance directly to local service providers, as was done in the Social Education Fund, while simultaneously strengthening provincial capacity, and increasing the expectations and demands by beneficiaries and clients for improved service quality. iv 1. CHAPTER BACKGROUND AND METHODOLOGY FOR THE STUDY CHAPTER SUMMARY The Federal Government in Argentina adopted important social sector reforms in the 1990's and provided assistance to the provincial governments responsible for implementing them. These reforms had some successes but so far they have not achieved fiully all their goals. Thus, there is still a ways to do. This study examines some of the constraints to the reforms particularly the institutional capacity at the provincial level. 1. As part of the study, we look at (a) the economic, political, and institutional context within which the reforms were implemented, (b) the design or content of the reforms; (c) the implementation strategy of the Federal Government; (d) and the capacity of provincial governments to manage the change. Three provinces-Catamarca, Cordoba, and Salta-are studied in depth, including surveying public officials, holding focus groups and interviews with key stakeholders, and reviewing A. INTRODUCTION 1.1 Argentina is study in contrasts. For a developing country, it has high per capita income-just under US$ 8,000 per capita, yet it has a high degree of unemployment and 29.4 percent of the population in poverty. It has undertaken major economic restructuring and rationalization of the public sector since 1990, yet there remains a very large social agenda to be fulfilled and many social sector institutions and services have avoided substantive restructuring. Education, health, and much of social assistance have been decentralized to the provincial governments, yet there is still duplication and fragmentation among levels of government, and many provinces are dependent on the Central Government for policymaking. Within the provinces both finance and decision- making remain highly centralized. Many complain that the reforms have yet to make a difference at the level of the school, hospital, clinic or other social service delivery agency. Democratic institutions and civil liberties appear to be healthy, yet there is little popular participation in government. 1.2 Another Argentine contrast concerns the effectiveness of its public institutions. Despite its relatively high income and educational levels, Argentina does not have a reputation for effective government. Using international indices, Figure 1.1 illustrates that government in Argentina is perceived as being significantly less effective than countries like Chile and Spain and, also, significantly less effective than would be predicted given its level of per capita income. 1 Figure. 1.1: Relation Between Government Effectiveness and GDP Per Capita, All Countries 120 100 ~~~~~~~~~~Spain Chile 4)~~~~~~~~~~~~4 (c 80~ 1>40 X 0 . Argentina *~~ *44. * '~~~Medco 20 - --, . ., ___ 6 7 8 9 10 11 GDP per capita Source: Kaufmann D., Kraay A. and Zoido-Lobaton P. (1999a) 1.3 In recent years the Federal Government has initiated a number of reforms to improve the effectiveness of its public institutions. The purpose of this study is to assess the implementation of selected social sector reforms with the objective of learning how success in implementation is affected by the economic and political context, the design of reforms, the implicit or explicit implementation strategy, and the capacity of government to deliver services. 1.4 This study examines the implementation of two reforms-one in education and one in health-in three of Argentina's twenty-three provinces plus the city of Buenos Aires. Since every reform and every province has its own unique characteristics, the implications of this study for the implementation of future reforms must be interpreted with caution. B. RATIONALE FOR THE STUDY 1.5 The delivery of social services in Argentina has been largely decentralized to the provincial governments. A significant share of the delivery of public health services, primary and secondary education, and social welfare programs are all mainly the responsibility of the provinces. While delivery and financing have been largely decentralized, the Federal Government is still an important source of finance for social protection, for the national education reform, and for strengthening social sector ministries in the provinces. While small as a share of total education and health finance, federal sector-specific financing can represent a large share of the monies available to provincial ministries for new investment. 2 1.6 Despite decentralization, the Federal Government retains the core responsibility for ensuring the provision of the public goods of the nation, especially, equity and economic growth. Ineffective and inefficient delivery of social services constrain the effectiveness of efforts to improve equity, alleviate poverty, and provide the human capital that underlies economic growth. The Federal Government, also, has a key responsibility for providing the information, knowledge, and enabling environment required for the efficient and effective delivery of social services. In keeping with these responsibilities, the Federal Government has attempted, in many cases with the financial support of the World Bank and the InterAmerican Development Bank, to stimulate reforms in the delivery of social services by the provinces. 1.7 Federal officials lament that their financing yields too little in terms of improvements in the delivery of social services. Federal monies are often fungible, and it is difficult to hold provincial governments fully accountable for their use. On the other hand, provincial officials view federal monies as coming with too many strings attached, which constrain the design of policies and programs appropriate to the particular province and which impede efficiency and effectiveness in the use of funds; and see reforms as being unfunded mandates which the provinces are expected to finance. And as shown in the interviews of 1,200 households in 29 cities conducted as background for the World Bank's recent report on poverty in Argentina, the poor view themselves as having little voice in influencing the delivery of social services.1 The poor, also, identify the National Government as the institution with the highest capacity and competence to respond to their needs, despite the fact that decentralization has handed much of this responsibility for providing services to the poor to the provincial governments. In addition, social sector reforms initiated at the Federal level and dependent on the provincial governments for implementation are widely viewed as not having had a large impact on service delivery. 1.8 The key role of the Federal Government in bringing about reform and change in the delivery of social services by the provinces, the perception by many officials and beneficiaries that Federal reform efforts yield meager results, the need to improve the design and implementation of reforms in order to bring about better social services to the poor and non-poor alike, and the limited knowledge at present in Argentina and elsewhere on how best to bring about lasting change in government provide the rationale for this study. C. ISSUES TO BE ADDRESSED 1.9 In line with the problem and rationale given above, this study attempts to answer the question what are the constraints to advancing further on the reform agenda and generate a greater impact on service deliver in the provinces.. To answer this question requires a careful examination of: * The context-institutional, political, and economic-within which reforms have been introduced and which influence provincial implementation. The World Bank ((2000), Argentina: Poor People in a Rich Country, Report No. 19992-AR. 3 * The design and content of the reforms themselves-their objectives, content, and actions. * The strategy-implicit or explicit-of the Federal Government ministry responsible for ensuring implementation by the provinces, including operationalization of the reform. * The key stakeholders and their views and interests in seeing the reform implemented. * The capacities of the provincial ministries and service delivery units to manage change. 1.10 Given the important role of the province in implementing social sector reforms in Argentina, this study puts emphasis on analyzing provincial implementation capacity. There is no single accepted framework for predicting how successful a ministry or other government body will be in implementing reform. There is by now a large research literature on the question of how to introduce and sustain change in the social sectors, but this literature draws on several disciplines, including political science, sociology, and economics, each of which has its own conceptual framework and terminology. The research literature, also, treats several distinct topics, including the political economy of change, organizational capacity to manage change, technical assistance to strengthen institutional capacity, and the institutional rules of the game that underlie any reform effort. Much of this literature is useful in suggesting variables that might influence the implementation of social reforms in Argentina . Unfortunately, there do not exist any good reviews of this literature which might provide a synthetic model to guide this study, and it is beyond the scope of the work here to do this review. Our approach is by necessity ad hoc in the sense that we have reviewed the existing empirical literature and grouped the factors that appear to most strongly affect implementation success into six key characteristics. * Leadership and clarity of goals: Do leaders offer a clear vision for the future? Effective organizations almost always have leaders who are able to create a sense of vision, mission and focused purpose. Whether governors, ministers, and other leaders choose to provide this leadership depends in part on the political and other rewards for doing so, i.e., the incentives they face. * Commitment and ownership: Are the ministry officials and other stakeholders responsible for the day to day implementation of the reform committed to it? Commitment reflects in part the extent to which key actors agree with the proposed reform, which is determined in part by their sense of ownership. There are potentially 2 Writings that have been especially helpful in this respect include: Bardach (1998), Burki and Perry (1998), Graham and Naim (1997), Grindle and Hilderbrand (1995), Grindle and Thomas (199 ), Haggard and Webb (1994), Hirschman (1963), and Manning, Mukherjee, and Gokcekus (2000), Morah (1996), North (1990), Orbach (2000), Pinto and Mrope (1995), Tendler (199 ) Tendler and Freedheim (1994), and Tommasi and Velasco (1996). 4 serious agency problems in social service delivery. Often, the staff responsible for providing the service can "hide" their real performance and, thus, have the capacity to "shirk". One means of reducing this risk is to make these staff the "owners" of the reform, both through effective communication to those staff, by "listening" to their concerns and taking steps to accommodate them, and by allowing them to participate to the extent possible in defining the reform and the implementation strategy. * Authority: Does the province, the ministry, or the service delivery unit have the tools and authority to take action? Delegation of responsibility is not always accompanied by the authority to allocate resources and to employ, evaluate, and discipline staff. * Incentives: Are there adequate and appropriate incentives? Incentives need not be monetary in nature, but they do need to establish a linkage between the contribution of the individual or the team to the meeting the organization's goals and valued rewards. * Accountability: Do mechanisms exist to ensure accountability to the client? Effective organizations are those that are held accountable for their performance, both to their financing sources and, more importantly, to their clients for the services provided. Accountability requires that staff or administrative units know the performance they are expected to achieve. It, also, requires transparent information regarding performance, available to client and provider alike. * Resources: Are there adequate human andfinancial resources? Are they stable and predictable? Implementation of reform requires additional resources, both human and financial resources, and their efficient use. Transparent and merit-based recruitment and promotion standards help ensure human resources are both adequate and used effectively. Predictable financial resources affect an organization's ability to plan and to execute its plan. D. METHODOLOGY OF THE STUDY 1.11 To answer the questions asked earlier, this study looks at two social sector reforms adopted in the 1990's: a comprehensive, national education reform approved by the Argentine Congress in 1993 and a more specific reform of public hospital management initiated on a pilot basis, also in 1993. These reforms are assessed in terms of their design, the implementation strategy adopted by the Federal Government, and their impact on service delivery, to the extent that is known. To more carefully assess how these reforms were in fact implemented at the provincial level, the study, also, selected three provinces-Catamarca, Cordoba, and Salta-where more detailed information was gathered from focus groups and interviews of important stakeholders, and particular emphasis was put on understanding the rules of the game of the provincial bureaucracy. 1.12 The Sample. The Argentine Republic is a constitutional federal republic created by the Constitution of 1853. The 23 provinces and the Municipality of the City of 5 Buenos Aires (with legal status similar to the provincial governments) enjoy substantial autonomy. They execute approximately 38 percent of total public sector expenditure in Argentina, and the 2,159 municipalities and towns in Argentina are responsible for another 7 percent of public sector spending. Each province has autonomy to organize its territory into local municipal jurisdictions. 1.13 For the purposes of this study, we have selected three provinces which, taken together, reflect the heterogeneity of the provinces outside the city and province of Buenos Aires. Both the city and province of Buenos Aires are significantly different than any other province. The city of Buenos Aires is far wealthier than any other province, with a GDP per capita of over $25 thousand per capita, more than three times the national average. The province of Buenos Aires is far larger than any other province, with a population of 14.2 million, more than nine times the size of the average province and 38 percent of the national total. 1.14 The three provinces selected for study are Catamarca, Cordoba, and Salta. Taken together, these three provinces represent diversity in terms of size, wealth, size of government, and politics. As will be seen later, they also demonstrate diversity in terms of the implementation of social sector reforms. Table 1.1: Characteristics of the Provinces Variable Catamarca Cordoba Salta Population 0.3 3.0 1.0 (millions) _ Percent Poor (NBI) 28.2 15.1 37.1 Per Capita Income, 5.0 7.9 4.5 1996 (thousands) Size of Government 81 28 44 a) Literacy Rate 95.0 96.5 92.3 Infant Mortality 25.6 16.3 20.5 Rate Political Party of Civic & Social Front Radical Justicialista Governor (1996-99) 1 _ a) Number of public sector employees per one thousand population 1.15 Catamarca. Catamarca is a small, poor province of the Argentine Northwest. Its per capita income is about half the national average, and is also ranks poorly on social indicators. More than one-fourth of the population cannot meet their basic needs, as measured by the Indicator of Unmet Basic Needs, or NBI, compared to a national rate of 18.9 percent. It has high but diminishing infant mortality rates and lower school attendance rates that are well below the national average. The net school enrollment rate for primary education is only 63 percent, suggesting high repetition and dropout rates. In addition, the quality of education in Catamarca is low, according to the national 6 assessment test. Catamarca shows a better performance in terms of literacy rates (closer to Cordoba than to Salta). 1.16 The province attains low social indicators despite per capita health expenditures equal to the national average and despite having high educational spending per pupil, relative to the national average. Catamarca is highly dependent on the public sector for employment, and provincial finances are highly dependent on Federal Governnent transfers. The province has been governed by the Frente Civico y Social, a coalition that was in the opposition to the national government during the period considered. 1.17 Cordoba. Cordoba is the third largest province in Argentina with more than four times the population of the average province outside Buenos Aires. While its per capita income lies just about the national average, its poverty rate as measured by the NBI population is better than the national average. Outside Buenos Aires, it has the third lowest NBI index in the country. Also, it has higher education and literacy rates and lower infant mortality rates than the nation as a whole. Cordoba's provincial government is relatively small in terms of public employment, and it generates a considerably higher percent of own-source revenues than the average province.3 1.18 Politically, Cordoba has a history of being in opposition to Buenos Aires and the national government. It is proud of its history and jealous of its autonomy. Among other achievements, it was the origin of the university autonomy movement in Latin America. In more recent times, its government has been governed by the Radical Party since the resumption of democracy in 1983 and, thus, was in opposition to President Menem during the 1990s. More recently, in 1999, it elected a Justicialista Party governor, while the country was electing a Radical Party President. 1. 1 9 Salta. Salta is a relatively poor province in Northern Argentina. In terms of size, it is slightly larger than the average province outside Buenos Aires. Its per capita income ranks 15th in Argentina and is less than half the country's average. It accounts for 1.3 percent of Argentina's GDP and 2.8 percent of the country's population. While its per capita income is not greatly different from that of Catamarca, it has a much higher percent of the population with unmet basic needs. Its NBI index is fourth highest in the country and almost double the national average. Its size of government as measured by public sector employees per thousand inhabitants is below the national average and about midway between Cordoba and Catamarca. 1.20 Salta is the only one of the three provinces which has had a governor of the same political party as the country's president during the period in which social sector reforms were being implemented in the 1990's. It is, also, the only one of the three provinces to participate in the World Bank financed Second Provincial Reform Loan. The Implementation Completion Report for that Loan gives Salta a highly satisfactory rating for implementation success and noted that Salta had "superior fiscal performance over the 3 For a detailed analysis of Cordoba's public sector, see E. Nofal (2000), Analisis del Sector Pz4blico de la Provincia de Cordoba, 1995-1999. 7 1997-99 period relative to other provincial fovernments in terms of own-source revenues, personnel expenditures, and stock of debt." 1.21 The Methodology. The methodology used to assess provincial experience in implementing education and health sector reforms includes analysis of the empirical evidence complemented by information obtained from surveys and focus groups held with stakeholders in each reform area. 1.22 The principal survey instrument used in this study was the Public Officials Survey developed by the World Bank with Dutch Government support to measure many of the variables indicated earlier. While this survey instrument has been applied in several countries in the world, it has been used in only one other country in Latin America- Bolivia.5 The instrument was modified for use in the Argentine provinces, in particular, by adding several questions concerning intergovernmental relations, and was administered to public officials in four ministries-Economy and Finance, Education, Health, and Social Welfare.6 In total, 370 randomly selected officials participated in the survey as follows: Catamarca (90), Cordoba (153), and Salta (127). The survey was carried out in November 1999. 1.23 In the case of education, World Bank experts in institutional development and Argentine education experts worked together to design three complementary survey instruments. The first instrument, Questionnaire for Officials in Provincial Education Ministries, was administered to 70 randomly selected public officials in the education ministries as follows: Catamarca (13), Cordoba (27), Salta (30). The second instrument, Questionnaire for School Directors, was administered to a randomly selected sample of directors of 38 primary schools as follows: Catamarca (12), Cordoba (15), and Salta (11).7 The third instrument, Questionnaire for School Inspectors and Supervisors, was administered to the school inspectors assigned to supervise the schools included in the sample for the second instrument. Given the small sample (15) of inspectors, this questionnaire called for open-ended rather than structured responses. 1.24 Given the small sample size used in the education sector questionnaires, responses were analyzed only in terms of average responses to questions.9 Logit analysis was used to estimate the relationships between the variables identified earlier and organizational effectiveness, as perceived by public officials. 1.25 To complement the data obtained from these survey instruments, focus group interviews were, also, held with diverse groups of stakeholders. These included social sector ministers, the heads of key departments, union officials, managers of service 4 World Bank (2000), Implementation Completion Report, Argentina: Second Provincial Reform-Salta, Report No. 20698. 5World Bank (2000), Bolivia. 6 The instrument itself can be found in the Technical Documentfor the Assessment of Institutional Capacity for Social Sector Reform in Argentina (the Technical Document), Winkler, Donald, 2001. 7Given the small sample and the limited travel budget, the sample was drawn from schools within four hours travel of the capital city. 8 All three education questionnaires are found in the Technical Document. 9Selected responses are reported in graphical form in the Technical Document. 8 delivery units, and, in the case of education, parents and school inspectors. In addition, officials in Federal Government social sector ministries were interviewed to obtain additional information on Federal implementation strategy and to obtain a Federal official perspective on provincial implementation performance. E. ROADMAP OF THE STUDY 1.26 To answer the questions posed earlier, this study first looks at the capacity of provincial governments to implement reform. Thus, Chapter 2 assesses the economic and political context of the provinces and how this context affects the capacity of provincial governments to implement reform. We then proceed to analyze the results of the Public Officials Survey to describe the institutional context and what effect this context has on the success that provincial ministries have in delivering services. 1.27 Given the implementation capacities of provincial governments, Chapter 3 looks at the implementation of the national education reform. We look at the respective roles of the Federal Ministry of Education, the provincial ministries, the school inspectors, and the schools themselves. Chapter 4 follows a similar organization in studying the implementation of hospital management reforms 1.28 Chapter 5 summarizes the lessons learned from the study and, for a hypothetical future reform, asks, "What kind of federal intervention strategy might work best in Argentina?" Special attention in this summary is paid to the relatively successful experience of the province of Salta in implementing the education and health reforms. 9 2. CHAPTER CHANGE AND REFORM IN THE SOCIAL SECTORS CHAPTER SUMMARY Among the factors affecting the implementation of reforms are the context in which the reforms are implemented, the design of the reforms themselves, and the capacity of provincial governments to implement new policies and programs. In this chapter we take a broad look at these factors, leaving more detailed analysis of specific reforms to subsequent chapters. The three provinces differed considerably in terms of context during the reform period. Fiscally, Catamarca was the weakest and Cordoba the strongest, and both Cordoba and Salta adopted policies to reduce fiscal deficits. Politically, Salta was the only one to have a governor of the same political party as the nation's president. In terms of overall provincial capacity to manage and implement policies and programs, civil service staff in all three provinces believe corruption is a serious problem, that political ties are important for career advancement; that ministries seldom have clear and consistent goals; that there are few rewards for exemplary performance or penalties for bad performance; that the performance standards required for clear accountability are generally lacking; and that financial resources are unpredictable. There are some regional differences, but overall provincial capacity is weak. In addition to the provincial differences, the education and health reforms studied here differ significantly in their formulation and design. In terms offormulation, hospital autonomy was the result of an executive decree, and had little visibility and relatively low support among health professionals. The education reform, in contrast, was the product of lengthy debate resulting in a Federal law and high visibility. In terms of design, provinces were mandated to implement several elements in the education reform, while hospital reforms were voluntary. In terms of Federal implementation strategy, the education ministry was more proactive than the health ministry in supporting reform, although both reforms received internationalfinance. A. DETERMINANTS OF CHANGE 2.1 The implementation of public policy reforms depends on many factors. First of all, reforms are initiated in a particular economic, financial, political, and social context. The economy may be growing, or it may be contracting, with increasing poverty and unemployment. The government's budget may be in balance, or it may be suffering from unsustainable current deficits and debt service. Political leaders may enjoy large parliamentary majorities, or they may have to reach compromises with their political foes. The reform may be initiated in a society where there is substantial equality or one rife with tensions, and possibly violence. Finally, reforms may be introduced in a culture 10 which values personal and political relationships highly, or one where individual performance is of paramount importance. 2.2 Second, social sector reforms are implemented by public bureaucracies. The capacity of these bureaucracies may vary widely not only in terms of human and financial resources, but, also, in terms of the rules of the game under which the bureaucracy functions. These rules of the game are less the formal rules laid down in staff and organizational manuals, which may or may not be read and enforced, than they are the informal rules that govern the daily behavior of those responsible for implementing the reform. For example, a bureaucracy where staff are evaluated and rewarded on the basis of their performance in accomplishing the mission and goals of the organization is quite different from one where there are no staff evaluations, and pay and promotions are based on political ties. 2.3 Third, social sector reforms can differ greatly in terms of their own design, and this, too, is likely to affect their implementation. Reforms can be comprehensive, or selective. They can be national in scope, or simply pilots from which one expects to learn. They can be designed in a highly participative fashion, or developed by a small group of technocrats. 2.4 In this chapter we, first, look at the context within which social sector reforms are implemented in Argentina. Second, we examine the capacity of provincial ministries to effectively deliver services and implement new initiatives. Third, we look at the designs of social sector reforms and how these designs accommodate the special challenges facing government in implementing reform. B. THE CONTEXT FOR REFORMS. 2.5 Social sector reforms in Argentina occur in the context of social norms and physical and human resources which play an important role in either facilitating or hindering their implementation. As noted earlier in Figure 1.1, government is in general perceived as being less effective than expected, given Argentina's per capita income. Among the factors which may contribute to this low level of effectiveness are corruption, which adversely affects incentives for performance, and a lack of citizen and beneficiary participation in government, which constrains the demand for reform and, also, reduces orientation and accountability to the client. 11 Figure 2.1: Relation between Control of Corruption and GDP per capita (PPP) All Countries 100 * Chile Spain * c70 40- * * t\~> 2entina 20 Mxc 6 7 8 9 10 11 GDP per capita (PPP) Source: Kaufmann D., Kraay A. and Zoido-Lobaton P. (1999a) Figure 2.2: Relation between Voice and Accountability and GDP per capita (PPP) All countries 100- Spain 90 70 60- 40 * * *, 4 30 4 ,e I * 6 7 8 9 10 1 1 GDP per capita (PPP) Source: Kaufinann D., Kraay A. and Zoido-Lobaton P. (1999a) 2.6 As shown in Figures 2.1 and 2.2, Argentina is perceived as having weak controls on corruption relative to its per capita income, and voice and accountability in government are significantly below OECD countries like Spain (but not significantly below what might be expected given Argentina's per capita income). Indeed, a simple model of government effectiveness predicts that if Argentina had the same level of perceived control of corruption as Spain, its government institutions would be just about as effective, too (Figure 2.3).1o 10 The model assumes that government effectiveness is a function of GDP per capita, controls on corruption, and voice and accountability and is estimated using a world sample of countries. 12 Figure 2.3 Actual and Predicted Government Effectiveness in Argentina 120 1 Argentina Spain 100 J ~~~~~~~~(predicted) 100 Chile c 80j * * * ; \M Argentina 20 * Me i; (actual) 6 7 8 9 10 11 | ~~~~~~~~~GOP per capita I ___________________ _ _ _ .__ . _ , , -- - ~ - -.-. __ _ _- ___ _ Source World Bank Estimates, Technical Document 2.7 Other variables likely to affect the implementation of reforms include the capacity to quickly communicate to the public to ensure their political support. As shown in Table 2.1, relative to other countries in LAC, Argentina enjoys a good communications infrastructure in terms of paved roads, access to telephones, coverage of daily newspapers, and the number of internet hosts. On the other hand, it lags behind Spain by a large amount with respect to both access to telephones and the number of internet hosts. Table 2.1 Several indicators 1998 Argentina Brazil Chile Mexico Spain Paved roads % 29.5 9.3 13.8 29.7 99.0 Normalized road index* 88.0 127.0 57.0 102.0 97.0 Telephone mainlines per 1,000 people 203.0 121.0 205.0 104.0 414.0 Dail news ers er 1,000 eople 123.0 40.0 98.0 97.0 100.0 Internet hosts per 10,000 people (1999) 27.9 18.5 21.5 23.0 76.8 *This index is an attempt to assess the adequacy of the stock of paved roads in a country at a given level of development. The value of the normalized road index shows whether a country's stock ofpaved roads exceeds orfalls short of the averagefor countries with similar characteristics. Source: 2,000 World Development Indicators. C. PROVINCIAL CONTEXT AND CAPACTIY 2.8 Economic Context. Government reforms occur in an economic context which affects greatly the prospects for successful implementation. As shown in Table 2.2, a relatively high percentage of provincial revenues come from national transfers, the largest component of which is the revenue sharing called Coparticipation. The financial dependence of the provinces on the Federal Government means their own finances suffer when Federal revenues are reduced for whatever reasons. When the Mexico peso was devalued in 1994, the "tequila crisis" adversely affected both economic growth and Federal revenues. The after effects of the Asian financial crisis and devaluation of the 13 Brazilian real in 1998-1999 had a similar effect. In both cases, reduced Federal revenues translated to reduced revenue sharing to the provinces. Table 2.2: Provincial Revenues and Debt Variable Catamarca Cordoba Salta Percent of Total Revenues from 81% 54% 69% Federal Government (1998) Provincial Own Source Revenues as 8% 36% 15% Percent of Total (1998) Federal Transfers Per Capita, 1995 $1041 $322 $454 Fiscal Deficit as Percent of Revenues, 16.5 19.1 15.6 1995__ _ _ _ _ _ Debt as Percent of Current Revenues, 66 41 60 1996 _____________________ Source: Ministerio de Economia [www.mecon.gov.ar/coord-pcias] 2.9 The amount of Federal transfers vary widely by province. As of 1995, Cordoba received the least as a percentage of its total revenues and on a per capita basis, while Catamarca received the most; this pattern remains true today. One of the most salient characteristics of many provincial governments in the 1990s has been fiscal deficits and high indebtedness accompanied by arrears in paying bills. External shocks, like the Tequila crisis or the Asian financial crisis, can serve to tilt a precarious financial situation into one demanding immediate adjustments, with consequences for the financing of social sector reforms. 2.10 Since the provinces have the primary role in financing and providing social services-they finance 97 percent of basic education, 92 percent of basic health, and 84 percent of social assistance expenditures-their fiscal difficulties have adversely affected the delivery of those services. Spending on investment and non-personnel inputs have been squeezed, and facilities have at times been closed due to work stoppages resulting from workers' frustration with wage and salary arrears. The pattern of provincial adjustment has differed across the three provinces studied here. 2.11 Catamarca's reform efforts in the 1990s began with a Federal intervention in response to fiscal disarray resulting from a combination of poor fiscal management and political turmoil stemming from a high profile criminal investigation. Subsequently, the province participated in two provincial reform projects financed by the World Bank. As a result of these efforts, fiscal deficits were reduced, the percent of current revenues spent on personnel declined. 2.12 Cordoba was affected especially severely by the Tequila Crisis, and, in response, a new political administration took drastic action. It cut civil service salaries by 30 percent, reduced public employment, and severely limited capital spending. As a result, a large fiscal deficit (19 percent of revenues in 1995) became a surplus by 1997. These actions, and the style in which they were implemented, created considerable friction with the public sector labor unions (Cavarozzi, 2000). 14 2.13 During the first half of the 1990's, Salta rapidly increased public employment and spending, resulting in a large fiscal deficit and growing stock of debt. Arrears in payments to employees and suppliers represented 35 percent of its total debt by the end of 1995. A new provincial government elected in 1995 embarked on a far-reaching and ambitious reform program aimed at reestablishing provincial fiscal balance, increasing tax collection, reducing current government expenditures and public employment, and reducing provincial debt. It quickly achieved a turnaround, paying its debt arrears and registering a fiscal surplus (including revenues from privatizations) in 1996. It, also, requested and received financial assistance from the World Bank as part of the Second Provincial Reform Project, approved in 1997. As a consequence of these efforts, the percent of current revenues spent on personnel declined (from 71 percent in 1995 to 55 percent in 1999), and own-source revenues increased. The World Bank's Implementation Completion Report for the Second Provincial Reform Project labels Salta "a clear success story." 2.14 While all three provinces were adversely affected by economic shocks, Salta responded relatively quickly and established a degree of stability and predictability in revenues and spending and did so in a way that ensured labor peace. Cordoba, also, took actions that reestablished an equilibrium between revenues and spending, but did so in a way that generated considerable labor opposition. Catamarca responded less quickly and, also, in a way that created conflicts with the public sector employee' union. 2.15 Political Context. The political context, also, affects the implementation of reforms. Since the resumption of democracy in Argentina in 1983, two political parties and two presidents have dominated national politics in the period 1983-1999. Raul Alfonsin, of the Radical Civic Union Party, was elected President in 1983 and served until 1989, when Carlos Menem, of the Justicialista (Peronist) Party, was elected. The 1853 Constitution, which has been altered several times, provides for a strong executive. Presidents appoint cabinet members and other executive officers without the need for legislative confirmation. The executive is empowered to submit legislation directly to the Congress, and approves or vetoes all legislation. Presidents also possess broad regulatory powers and may suspend civil liberties by declaring a state of siege. 2.16 Precedent has further expanded the statutory powers of the Argentine executive. Presidents have interpreted their veto power to include a partial veto of a bill, and may thereby redesign congressional legislation. A more conspicuous appropriation of legislative authority is the president's ability to enact decrees in cases of "necessity and urgency." Relative to other Argentine presidents, President Menem used this authority frequently, issuing 336 decree-laws between 1989 and 1994 alone. While Congress has the power to repeal decrees, the fact that President Menem enjoyed an absolute majority in the upper house (Senate) and a plurality in the lower house (Chamber of Deputies) meant few decrees were ever partially or totally repealed (Ferreira Rubio and Goretti, 1998). Support in Congress is especially important given that Argentina's electoral laws enforce party discipline. 2.17 It's reasonable to expect that Federal-provincial negotiations may be facilitated when the president and the governor are of the same political party. As shown in Table 15 2.3, of the three provinces studied here only Salta elected a governor of the same party as the president. Catamarca was and continues to be governed by a Radical-led coalition party, while Cordoba had a Radical governor until the 1999 elections resulted in a victory for the Justicialista candidate. Table 2.3: Political Indicators for the Provinces, 1999 Variable Catamarca Cordoba Salta Party of Governor prior to Civic & Social Front Radical Justice 1999 Does Governor's Party Yes Yes After 1996 have Majority in Parliament? Number of Years 8 16 16 Governor's Party in Power Results of 1999 Elections: Civic & Social Front Justicialista Justicialista Party of Governor I_I_I _ _ 2.18 To the extent having a legislative majority makes it easier to enact legislation, Catamarca and Cordoba were in the strongest position politically until 1996. Salta had the advantage of a close political affiliation with the Federal Government. Salta may have had the most politically advantageous position, having elected a Justicialista party candidate as governor in 1995 who had previously served in the national Senate and was well attuned to the workings of the Federal bureaucracy and, also, to the policy initiatives of the Menem administration. 2.19 Provincial Capacity. The finance, management, and delivery of social services in Argentina is highly centralized at the level of the province, and thus the capacity of provincial social sector governments is critical to success in implementing reforms. In this section we look at the capacities of these governments, as reported by their own employees and verified through interviews and focus groups. While it would be desirable to complement self-reports with more objective measures, this information is seldom available. The analysis is organized around the six questions posed earlier: * Do leaders offer a clear vision for the future? * Are the ministry officials responsible for the day to day implementation of the reform committed to it? * Does the province have the tools and the authority required to take action? * Are there adequate and appropriate incentives? * Do mechanisms exist to ensure accountability to the client? * Are there adequate andpredictable resources? 16 2.20 In addition to reporting the responses of provincial ministry officials to these questions, we, also, attempt whenever possible to look at the simple, two-way relationship between these variables and staff's self-assessment of their organization's effectiveness. Given the many caveats which must be made about the use of this measure of an organization's effectiveness, the results reported here must be interpreted judiciously. 2.21 Self-Assessment of Effectiveness. Provincial ministry officials do not, in general, rate highly the performance of their organizations, and the differences between provinces are relatively small. Officials in Cordoba rate themselves the highest, but even in this province only 27 percent of respondents agree that their ministry has been "highly successful" in fulfilling its mission objectives. As shown in Figure 2.4, among ministries, officials in the education ministry rate themselves the highest, and officials in social development and health rate themselves much lower. Figure 2.4: How Successful is the Ministry in Fulfilling its Mission Objectives? Percent Responding "Highly Successful" 40%- 35%- 34% 30% 26% 25% - 20% - X9%19% 15% - 0% ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~7 Argentina Ministry of Ministry of Ministry of Ministry of Finance & Education Social Health Economy Development 2.22 The reasons for the differences between ministries may partly stem from the fact that the Education Reform has given officials clear objectives against which to evaluate themselves and, as shown later, provincial education ministries have had some success in both increasing enrollments and provincial education budgets in recent years. As shown earlier (Figure 2.3), corruption may, also, affect the effectiveness of government, and provincial officials view corruption as a pervasive problem affecting all ministries and all provinces. Sixty-two percent of all respondents view corruption as a serious problem in their ministry. Seventy-nine percent also view political influence as a serious problem in their ministry. 2.23 Vision and commitment. One might expect greater clarity of and agreement with goals in provinces that have leaders with clear, focused agendas; in ministries with highly 17 visible sector goals; in ministries where employees participate in decision making, and in ministries where policy pronouncements are credible. And if all these factors are present, one would expect a relatively high degree of goals and policy consensus within the ministries. In fact, only 30 percent of provincial officials perceive their ministries as having clear and consistent goals (Q48), suggesting that employees may not feel a commitment to a clear vision or agenda. The responses of provincial officials provide evidence of their disagreement with ministry policies, of their lack of participation in ministry decision making, and of low policy credibility. 2.24 Further evidence on the lack of commitment to a clear vision is provided in the reports by ministry officials that they are often in disagreement with official policies. One explanation for the low consensus overall may be the hierarchical decision making style of the ministries. Of all respondents, 24 percent report they "almost never discuss policies and performance with colleagues in their ministry (Q53)". Both the consistency of policies and the degree of consensus within the ministry around those policies are correlated with staff perceptions of the effectiveness of their ministries, as shown in Figure 2.5. Figure 2.5: Consistent Policies and Internal Communication and Ministry Effectiveness 35% 25% - 315%[ 100/ Vey consistent NOT very Discusses Almost never policies consistent policies with policies others 2.25 The ability of ministries to implement their programs also depends on their credibility. Bureaucracies which are more oriented to satisfying political constituencies than attaining organizational goals are less credible and less likely to engage the energies of their employees. Transparent, merit-based recruitment practices contribute to credibility. Provincial ministry staff in general report that recruitment is not merit-based, and it is rarely openly advertised. Only 7 percent of respondents state they found their current position through a published advertisement (Q9). Only 22 percent claim positions like theirs are generally advertised to the public (QI0). And only 18 percent believe that employees in their ministry are hired primarily on the basis of merit (Q15). As shown in Figure 2.6, both public advertising and merit-based recruitment are positively associated with the perceived effectiveness of the ministry in meeting its goals. 18 Figure 2.6: Open and Merit-Based Recruitment and Fulfilling the Ministry's Mission Successfully 25% 4O%j 7n ~~~~~~~~20%~ 30% Hi g57y 15%; 0% .,., Merit-based NOT Merit- Advertised NOT-Advertised recruitmnent based recruitment 2.26 Authority. For a provincial ministry to be effective it must have the tools and the authority to implement its vision and mission, and its staff must feel similarly empowered to take action. In general, the provinces enjoy a great deal of authority in designing policies and programs and organizing themselves to implement those policies and programs. However, they also face important constraints, especially in terms of financing and labor relations. The fiscal status of the province, which is partly determined by the province's share of national coparticipation revenues, constrains ministry spending, and provincial labor regulations may constrain any particular ministry's ability to reallocate labor consistent with new goals. While new labor legislation and implementing regulations can be introduced, they may require considerable time to negotiate with public employees' unions and may require that the provincial governor expend considerable political capital to enact, even if the governor enjoys parliamentary majorities. 2.27 For provincial officials to take action, they must perceive that they have the authority to do so. Almost half of all respondents in the three provinces state that they feel they do not "have the appropriate amount of discretion in the performance of their duties (Q28)". This perception is consistent with others indicating that the typical provincial ministry is strongly hierarchical in nature. 2.28 Incentives. The incentives for ministry staff to align their own actions with ministry goals also affect the ministry's effectiveness. While most staff (81 percent) report their job performance has been evaluated within the past five years (Q22), only 39 percent report that evaluation was fair (Q24), 43 percent believe that staff with good connections to political leaders or powerful family members fare much better than other employees (Q62), 83 percent report that employees who are competent and hard working are not treated much better than those who aren't (Q63), 74 percent report that no staff have been disciplined for poor performance in the past three years (Q64), and 79 percent report that in the past three years no staff have been recognized or rewarded for "taking prompt action on citizen requests or complaints" or for "providing outstanding service" (Q67). In short, the perception of ministry employees is that good performance is not rewarded, while good political actions are rewarded. The incentives to help the ministry attain its goals are weak, excepting in those situations where the political elite and 19 powerful families in the province view these goals as important, too. As shown in Figure 2.7 below, there is a positive, and statistically significant relationship between an index of staff perceptions of the extent to which salaries and promotions are merit-based and the probability that officials perceive their ministry as being highly successful in attaining its mission." Figure 2.7: Perceptions of Merit-Based Rewards and Fulfilling the Ministry's Mission Successfully 30% 25% 20% 10% 5% 0% Merit-based reward NOT merit-based reward 2.29 Accountability. There are few mechanisms to ensure accountability of provincial ministries to their clients, or their sources of finance. Information usually flows downwards, from the ministry to the client-the school, the hospital or health clinic, the NGO or local government delivering food-and there is no systematic mean of obtaining feedback from clients on the services of the ministry. Neither is there much potential for client or citizen groups to undertake independent evaluations of those services-even ministry staff find it difficult to obtain even the most basic information about the ministry. For example, while 82 percent of ministry officials affirm that "my organization has financial records from the last five years, (Q30)" 37 percent state "it is difficult or very difficult to obtain information from those records (Q31)." In addition, there is no independent or semi-independent government body in the provinces that systematically audits or assesses the performance of the ministries and then makes the results known to executive and legislative bodies and the public. 2.30 Accountability also requires clear performance standards. Few public officials believe there are clear standards, with only 23 percent reporting "My organization maintains written standards of expected performance for its personnel (Q25)." Furthermore, as shown in Figure 2.8, there is a clear relationship between responses to this question and staff assessments of their own organization's effectiveness. 11 The index is comprised of two responses to two questions: (1) Have staff been recognized for taking prompt action on citizen requests or complaints and (2) Were these rewards justified? 20 Figure 2.8: Clear Performance Standards and Fulfilling the Ministry's Mission Successfully 40% 30% Highly 20% 10%- 0% , Maintain written NOT maintain written standarts standarts 2.31 Financial Resources. Adequate and predictable financial resources also contribute to an organization's effectiveness. Inadequate resources can prevent an organization from accomplishing its mission, especially if that mission includes major policy reforms. Unpredictable resources constrain an organization's ability to plan for the future and, thus, its capacity to set realistic goals and expectations both for the organization and for units and staff within the organization. High percentages of respondents in all three provinces report that actual funds for their ministry tend to fall short of budgeted funds, and that the size of the shortfall is relatively large. 2.32 Determinants of Ministry Effectiveness. Using the responses from the 370 provincial ministry officials in the three provinces of Catamarca, Cordoba, Salta, we estimate a simple logit model of ministry effectiveness, measuring effectiveness by official self-assessments and measuring the determinants by the responses that officials gave to the specific questions reported above. Reports by ministry officials are not necessarily accurate measures of effectiveness and capacity, and validating the reports of officials with independent measures of effectiveness and capacity was beyond the scope of this study. Hence, the results of this exercise should be interpreted with care. 2.33 In Table 2.4 below we report the results for those determinants of capacity which were statistically significant at the 0.05 level or better. Each determinant was also measured discretely and, thus, the impact reported in the table indicates the change in the probability that the ministry's mission would be accomplished very successfully. For example, advertising ministry positions when recruiting staff would increase the probability of a ministry accomplishing its mission very successfully by 11 percent, as contrasted with the alternative of not advertising ministry positions. The other variables found to be significantly related to mission success are having written performance standards, merit based staff rewards, retrievable records, consistency in policies and lack of political interference on the day to day workings of the ministry. A caveat: a more 21 complete model of organizational effectiveness would very likely alter the magnitude of the estimates reported in Table 2.4. Table 2.4: Determinants of Perceived Effectiveness and Their Impacts Detenninants Estimated Impact Advertising the positions 11.0% Having written performance standards 7.8% Merit-based reward system 6.9% Retrievable record keeping 16.0% Policy Consistency 17.3% No Political interference 6.9% Based on one-sided t-test results, only the coefficients--which are at least statistically significant at 0.05 level, are presented in this table.'2 2.34 The overall impact of the determinants of capacity on the ministry accomplishing its mission very successfully can be seen another way, Figure 2.9 reports the results of an exercise that calculates the probability of fulfilling the ministry's mission objectives successfully (our proxy for effectiveness) following a 10 percent incremental but uniform improvement in the capacity measures. The overall effects of raising capacity are large- for example, a 70 percent improvement in capacity roughly doubles the probability of fulfilling the ministry's mission objectives successfully. Given the caveats with the measures of effectiveness and capacity used in this analysis, these results should only be taken as suggestive of the magnitude of the impact of raising capacity. For more details, see the Technical Document. 22 Figure 2.9: The Potential Effects of Incremental but Uniform Improvements in all Aspects of Capacity. Probability(mission accomplished=Very High) 40%}- 25% 20% 15%- 5%- - 40% 1. 30%o -'- -- - - - -= = - initial |I level 10% 20% 30% 40% 50% 1 60% 70% 80% 90% 100% *argentia 14% 15% 17% 19% 21% 23% 125%t28% 32% 37% 43% D. SOCIAL SECTOR REFORMS IN ARGENTINA 2.35 Characteristics of Reforms. This study attempts to assess the implementation of two reforms in education and health and draw out the lessons to be learned for the implementation of future reforms in these and other sectors. The context of these reforms and the capacity of provincial governments to implement them play a role in explaining their impact on service delivery, but the design of the reforms themselves are also important. As shown in Table 2.5 below, the reforms being assessed have very important differences. 23 Table 2.5: Characteristics of Health and Education Reforms Characteristic Education Health Nature of Reform Comprehensive Pilot National Consensus on High Low the Problem Consensus by Key High Low Stakeholders on Design Popular Participation in Some None Reform Design Federal Law Yes No National Mandates Yes No Public Visibility High Low Transparent Outcomes Yes ? Financial Incentives Yes Yes Federal Ministry Proactive Passive Technical Assistance Yes Yes International Finance Yes Yes 2.36 The national Education Reform was the result of a lengthy national debate that produced national legislation that was comprehensive in its scope. Officials and stakeholders from all provinces thus had a degree of ownership of and commitment to the Reform. While some provinces might object to the implementation strategy of the Federal Government, they did not in general disagree with the Reform goals. The national legislation made mandatory the implementation of those key elements in the Reformn which could be easily monitored. Provinces might take actions to expedite or delay implementation, but they faced implementation deadlines that have the force of law behind them. The debate leading up to the Reform ensured that it had high visibility with provincial legislators, key stakeholders, and the lay public. Implementation mandates were, also, accompanied by financial and technical assistance provided by the Federal Ministry of Education and Culture, and this assistance was financed in part by international organizations which as part of their normal loan supervision require careful monitoring of the reformn's implementation. 2.37 The health reform reviewed here is the introduction of hospital autonomy in management of personnel and budget. This reform was not comprehensive, was voluntary on the part of the provinces, and was not the result of lengthy national debate. The reform had little visibility with the provinces, and the health care community at large, and neither felt much ownership of it. While the reform had financial backing from the World Bank, the Ministry of Health was not highly proactive in its implementation. 2.38 Table 2.5 summarizes the key differences between the education and health reforms. As shown, they differ in terms of design, process of formulation, and in Federal implementation strategy. The process leading to the education reform was more participatory and visible than was that of the health reform and resulted in legislation 24 approved by Congress, rather than an executive decree. The education reform had the further benefits of national mandates, as opposed to voluntary compliance, and transparent outcomes in terms of enrollment and learning. 25 3. CHAPTER EDUCATION: PROVINCIAL EXPERIENCE IN IMPLEMENTING A COMPREHENSIVE NATIONAL REFORM CHAPTER SUMMARY Following years of national debate and an important Pedagogic Congress, the Argentine national Congress passed legislation reforming education in 1993. While the responsibility for implementing the reform lay primarily with the provinces, the Federal Ministry of Education took a decidedly proactive role in guiding and assisting implementation efforts. This included seeking financial support from both the InterAmerican Development Bank and the World Bank. This chapter looks at the background leading up to the reform, assesses the Federal implementation strategy, analyzes provincial capacity to manage education and educational change, and reviews the role and capacity of the school to implement internal changes We conclude that the reform effort had several successes, especially in terms of achieving goals of increasing compulsory schooling and the government finance required to cover increased enrollments. However, for a variety of reasons, the reform was less successful in achieving the goals of changing classroom practices and raising qualitv. A. EDUCATIONAL REFORM 3.1 Argentina's 1993 Education Reform was a major, comprehensive national effort to improve primary and secondary education. Since the details of this Reform have been described and evaluated in another Bank study (Experton, 1999), the discussion provided here is brief. In what follows we, first, describe educational attainment in Argentina relative to international comparators and education attainment in the provinces studied relative to their comparators. Second, we report the objectives and actions or design of the Reform. Third, we describe the Federal Government's strategy for implementing the Reform during the 1993-1999 time period. 3.2 Context of the Reform. The current Argentine education reform has its roots in 1976, with the transfer of primary schools from the national to provincial governments, and the economic crisis of the 1980s, which was generally perceived by educators as adversely affecting education. The average monthly teacher salary, for example, decreased from $1,330 in 1984 to $619 in 1990 (IIPE, 1998). By the 1990's, Argentina was continuing to outperform most other countries in Latin America in educational attainment but had not yet reached the levels of most OECD countries. As shown in Figure 4.1, Argentina has an expected educational attainment for the current generation of school children in Argentina that is slightly lower than expected for its income level and is well behind higher income OECD countries. No data are available to compare the quality of Argentina education with that of OECD countries. As shown in Figure 3.2, Argentina outperforms most other Latin American countries, but Latin American countries' learning achievement in general is low, given their performance on the tests 26 administered by the International Education Association. (To date, Argentina has not participated in these intemational tests, although it did participate in a Latin American test administered by the regional UNESCO office.) For purposes of comparison, the predicted performance of Spain is reported in Figure 3.2.13 Figure 3.1: Years of Educational Attainment and GDP Per Capita 25 Spain 20- C '. Mexico 0 10 Argentina w 5 Chile 6 7 8 9 10 11 GDP per capita Source: OECD Education Database. Expected schooling (in years) under current conditions estimated by OECD. (education for children under the age of five is excluded). Figure 3.2: Third Grade Language Achievement and GDP Per Capita, Latin America 1320 - 300 -- , 7 280 -Chil 260 Argentina 240 220 Mexico 200 _ . 7 7.5 8 8.5 9 9.5 10 GDP per capita Source: UNESCO data in ClNDE/Dialogo Interamericano, 1999. * estimated 13 The exercise consist of assuming that the ratio of the median Spain score to the median Colombia score in the IEA test predicts how Spain would have performed in the OREALC test. The prediction is that it would have scored a median of 302 and a top quartile score of 333, higher than Argentina's actual median of 251 and top quartile score of 278. 27 3.3 Within Argentina, there are, also, large variations in educational performance. Figures 3.3 and 3.4 show the average scores on tests of language and mathematics achievement in 1998 as related to per pupil expenditure. The rank ordering among the three provinces studied here are consistent, from high to low: Cordoba, Salta and Catamarca. Both Salta and Catamarca under perform relative to what would be expected given per pupil expenditures. However, regressing achievement on per capita income shows that Salta performs better than expected given its per capita income. Figure 3.3: Relation between Seventh Grade Language Achievement Test Scores and Education Expenditure per Student, 1998 80 1Cordoba National 70- ~65- _ _ _ _ _ _ _ _ _ _ _ _ e 60; -+ * Salta s0 Catamarca 45 40 -- 0 500 1,000 1,500 2,000 2,500 3,000 3,500 Education Expenditure per Student Source: Direcci6n Nacional de Evaluaci6n de Calidad, Ministerio de Cultura y Educaci6n Direccion Nacional de Programacion del Gasto Social Figure 3.4: Relation between Seventh Grade Mathematics Achievement Test Scores and Education Expenditure per Student, 1998 75 - 70 Cordoba National 65- I ~60~ 8 2 to55-4 * -0 Salta 45 Catanmarca 0 500 1,000 1,5'00 2,000 2,500 3,000 3,500 Expenditure per Student Source: Direcci6n Nacional de Evaluaci6n de Calidad, Ministerio de Cultura y Educaci6n. Direcci6n Nacional de Programaci6n del Gasto Social 28 3.4 A growing call for education reform resulted in the 1988 Pedagogic Congress in Cordoba, in which most of the important stakeholders participated. This Congress in turn provided the basis for the 1993 education reform, approved by the National Congress in the Federal Education Law, which replaced the 19 century General Education Law. Another important and unexpected event occurred in 1991 when the Federal Government legally transferred its secondary schools to the provinces, more for fiscal than educational reasons. Finally, the National Congress, also, approved complementary legislation, the 1995 Federal Education Pact, for the purposes of improving provincial capacity, teacher training, strengthening local school management and participation, and improve school infrastructure. 3.5 Objectives and Actions. The 1993 Reform had as objectives increasing educational attainment, raising its quality, and improving equity. The Federal Education Law mandated a national increase in compulsory schooling from seven to ten years according to a specified time schedule. To raise quality, it changed the structure and curriculum of both primary and secondary schooling; established a national student evaluation system, and gave schools and local education communities increased authority to design and implement school development plans. To improve equity, it established the Social Education Plan which, among other things, sent monies directly to targeted schools, which were authorized to manage the monies and implement their own projects. Of the three goals of the Reform, only the first set explicit results-oriented targets. The quality goal was specified in terms of inputs [e.g., curriculum change] not improvements in learning [e.g., gains in test scores], and the equity goal was, also, specified in terms of inputs [additional resources] not outputs [reductions in disparities in learning]. 3.6 Implementation Strategy. Three key actors have had responsibility for implementing the Reform: the Federal Government, the provincial governments, and the schools. The national legislation creating the Reform gave the Federal Government lead responsibility for implementation. The Federal Government's implementation strategy for the Reform had four key components: (1) using a federal mechanism-the Federal Education Council--for arriving at consensus among the provinces as to policies and programs; (2) providing financial assistance to the provinces via the Social Education Plan and the Federal Pact; (3) offering technical assistance to provincial ministries of education (e.g., the PREGASE program to strengthen provincial ministry management) and (4) communicating information to teachers, directors, and other educational personnel through various means, including a monthly education magazine. 3.7 The provincial governments had responsibility for financing the required additional expenditures; employing new teachers and constructing new classrooms; adapting Reform elements to provincial needs and goals; setting a provincial timeline for implementation; and assisting schools and teachers to make the changes required to keep to that time line. The schools-teachers and directors-had the responsibility to make the required changes in school organization and curriculum and to define their own priorities and needs and solicit funding to meet them. 29 B. IMPLEMENTATION OF THE REFORM 3.8 The Education Reform had an impact on educational outcomes, but some provinces did better than others. Given the design of the Reform, the change in educational outcomes can be viewed as dependent on four major elements: (a) the support or opposition to core elements of the Reform by the key education stakeholders in the provinces, (b) the success of the Federal implementation strategy, (c) the capacity of the provincial education ministries to manage their educational systems, and (d) the capacity of schools to implement change. After describing the changes in educational outcomes in the provinces during the reform period, each of these elements is considered in turn. 3.9 Educational Outcomes: Provincial Implementation Experience. As shown in Table 3.1 and Figures 3.5 and 3.6, during the reform years the provinces increased educational spending, expanded coverage of secondary education, and raised average test scores. Figure 3.5, for example, shows that secondary school enrollments increased faster than population in almost all the provinces, although the size of the increase varied widely. While these results cannot be entirely attributed to the reform effort itself, they suggest progress towards the reform's overall goals.'4 Among the three provinces studied, Salta shows the largest gains in test performance, despite having the lowest per pupil expenditure (Figure 3.3 above). It, also, shows the largest gain in secondary school gross enrollment rates. Table 3.1. Indicators of Provincial Implementation of Education Reform Indicator National Catamarca Cordoba Salta Average Absolute Change in Secondary +10 +13 +14 +15 School Gross Enrollment Rate [%], 1991-1996 Change in Public Expenditures +1 +1 +1 +3 on Education as Share [%]of Total Provincial Government Expenditures, 1993-1998 Absolute Change in Average 7th +4.2 +5.3 +1.5 +8.6 Grade Test Scores (Mathematics + Language), 1993-1999 __f Source: Ministry of Education, Morduchowicz (1999) 14 The magnitude of the changes in test scores changes significantly depending on the years one selects for comparison, perhaps reflecting the changes in the sample of schools tested from year to year. Nonetheless, the pattern of changes across provinces does not change significantly with the selection of years. 30 Figure 3.5: Percent Changes Between Population and Secondary School Enrollments, 1994 - 1998 K40.001 . 35.00 at E 35 02_'alta 3 30.00 - WU National 25.00 *P to 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 L_____- % Changes in Population Source: Direccion Nacional de Evaluacion de Calidad, Ministerio de Cultura y Educacion l)irecci6n Nacional de Prograrnacion del Gasto Social Figure 3.6: Average Language Achievement Test Scores Grade 7, 1994-1998 085.00-- ---- - - 80.00 - 7.00 1 70.00-S o65.00 - Cordoba \* < Cordoba M ~~~Catamarca// 40.00 - 40.00 45.00 50.00 55.00 60.00 65.00 70.00 75.00 80.00 85.00 Teat Scores 1994 Source: Direccion Nacional de Evaluacion de Calidad, Ministerio de Cultura y Educaci6n Direcci6n Nacional de Programaci6n del Gasto Social 3.10 The Role of Key Stakeholders in Implementing Reform. The implementation of education reform in the provinces involves many actors, who do not all share the same interests and objectives. Part of the task facing the provincial ministry is the construction of consensus around the reform and its implementation, both within provincial government itself and among the various stakeholders who may either facilitate or pose 31 obstacles to implementation. In what follows, we briefly describe the major stakeholders and their views on key elements of the Reform. 15 The stakeholders considered here are: * The provincial governor. * Ministry senior management. * Supervisors and inspectors. * Teacher union leadership. * School directors. * Teachers. * School cooperatives and parents. 3.11 Provincial governors play a major role in setting provincial government priorities, as is often reflected in the allocations of provincial finances and in the people named by the governor to important political positions, in particular, the education minister. In most political economy analyses of education reform, the costs, most of which are incurred in the short-run, exceed the benefits, most of which accrue in the long-run. However, two factors in modem day Argentina may change this calculation. First, the quantitative goals of increased compulsory schooling were legally mandated. Second, improving education has become a politically more popular issue as open markets and new technology have put higher premiums on well-educated labor. A governor who wishes to be seen as a modern, progressive reformer can point to educational improvements as part of his political legacy. 3.12 Ministry senior management includes the minister and heads of departments, almost all of whom are typically political appointees of the governor. The role of provincial education ministers in implementing the reform has been critical, both in constructing consensus in the Federal Education Council and in convincing the provincial political authorities to provide the extra financing required to construct the extra schools and employ the extra teachers needed. Also, if provincial capacity in managing education is to be strengthened, the minister has to be a champion for change. Finally, it is up to the minister to develop provincial policies and programs that are consistent with and reinforce the national reform. 3.13 Supervisors/inspectors. Supervisors or inspectors are the face of the provincial ministry to school directors and teachers and in a unique position to communicate the goals and programs of the reform and to assist schools in their implementation. Their incentives may differ depending on their appointment, which may be political (Salta) or governed by strict civil service rules (Cordoba). Salta offers the advantage of inspectors who share the minister's vision, while Cordoba offers the advantage, in principle, of strong professional qualifications and long term working relationships with the schools. In practice, civil service competitions for inspector have not been held in recent years, so most inspectors are temporary and have been selected on the basis of longevity. 15 More detail is provided in the Technical Document. 32 3.14 Teacher union leadership. The union leadership is democratically elected and, thus, must respond to their clientele, the teachers. While the unions support the Reform in principle, they find that in practice the Reform has imposed costs on teachers. Also, in most provinces, the Reform was implemented at the same time that fiscal crises led to reductions in teacher pay or late payments. As a result, all three provinces studied here suffered from work stoppages, although in Salta's case labor stability subsequently resulted from a firm provincial commitment to pay teachers on time. 3.15 School directors Directors have little authority and a great deal of responsibility for providing leadership in implementing change. In addition, a significant percentage hold temporary appointments, and may revert to the teacher corps once civil service competitions for the position are held. Focus groups held with directors of primary and secondary schools in the three provinces studied reveal a high degree of commitment to implementing the reform, including the school development plan, and an equally high degree of frustration with their ability to manage resources to make this possible. Directors skilled in the art of persuasion may create commitment on the part of teachers to the reform. 3.16 Teachers. We did not survey or systematically interview teachers as part of this study, but previous surveys (e.g., IIPE, 1998) show them to be unenthusiastic supporters of reform. They view reform as an imposition from above and its implementation as following a political rather than technical timetable. In addition, they view the reform as personally costly, both because it has imposed retraining requirements on them and, also, because fiscal deficits in several provinces resulted in reduced pay or job instability at the same time the reform was being implemented. 3.17 School Cooperatives and Parents. As part of this study, we held focus groups with the parents comprising school cooperatives in the three provinces studied. The results of these focus groups, as well as questions asked of other stakeholders regarding parents, reveal a strong interest to be more involved in school decision making and, also, a strong resistance on the part of school and ministry officials to the same. An earlier assessment of parents in Buenos Aires Province (done for another study) shows that they have little knowledge of the reform and believe they have no role in implementing it. This, combined with their perception that their participation is not welcome, explains in part the fact that school directors report parents show little interest in participating school development projects (Q2.2). 3.18 Views on Specific Reform Elements. As described earlier, the Argentina education reform has many elements, which have been financed and implemented unevenly. The views of key stakeholders with respect to three of the most visible elements were sought during the focus groups held with each. Those elements are (a) school development projects (SDP), (b) the national student assessment, (c) and in- service teacher training. Their views are summarized in Table 3.2 and in the comments below. 33 3.19 School Development Projects. There is a high degree of consensus among stakeholders as to the importance of the SDP. The only dissenting view comes from parents, who believe the SDP process is not sufficiently participatory. However, the importance assigned to the SDP appears inconsistent with the finance and training provided. Only 25 percent of school directors report receiving government financial support for the SDP (Q2.4). A study of the experience with the SDP in Cordoba confirms that schools lack the training, guidance, and financing needed to realize the potential of the SDP (IIPE, 2000). Table 3.2: Stakeholders Views of Reform Elements Stakeholders School Development Projects National Student Assessment In-Service Teacher Training Education Ministry Implementation of the SDP is The national assessment is of A very high priority for improving Officials critical to advancing the reform. limited relevance; high priority quality, but national training is of is put on developing aprovincial limited utility. Training provided assessment system. under the Social Pln was excellent. Supervisors/Inspect Guidance on and monitoring of Assessment results help in More training is needed, and more of ors the SDP has become a key part identifying schools with special it needs to take place in the school. of the supervisor's job. needs. Teacher Union The SDP won't be a useful Teachers are at risk of being Teacher training is good. The quality Leadership management tool until teachers, blamed for low scores. The of training provided by the union is directors, and inspectors are national assessment is not high. better trained in its use. aligned with provincial learning goals. School Directors The SDP is fundamental to Test results help directors to get Training needs to be adapted to the school development and to teachers to focus on learning needs of the school. Social Plan improving quality, and to problems. All schools shouldbe training was very good. The province involving parents. tested each year, and schools under invests in training. should be helped to design remedial actions. Teachers The burden of preparing the Tests would be more helpful if Teachers should be given time for SDP falls on the teachers; they gave teacher informaton on training. parents do not want to be the performance of individual involved. students. School Only a few committed teachers Parents need to be educated as to There should be traning workshops Cooperatives & prepare the SDP. Despite the the meaning of test results. for parents, so they can participate Parents talk, parental involvement is not more fully in the school, especially in _ welcomed. the SDP. 3.20 Student Evaluation. While Argentina quickly implemented a national student evaluation system, there is discord among stakeholders, and among provinces, too, as to its value. Ministry officials are supportive when their province does well relative to the national average, but the overall average level of support is low (Q6.8). Officials uniformly complain that the results are not well disseminated (overlooking their own role in this). Some provinces seek to develop their own assessment systems which more specifically measure provincial goals. Union officials are sometimes fearful that teachers could be blamed, when school performnance is poor. Inspectors and school directors approve of national student evaluation and, in general, strongly encourage that it be expanded from a sample to a census basis covering all schools and, especially, that it be extended to provide specific advice as to what schools need to do to improve. Finally, parents generally express ignorance of the evaluation system and its results for their own schools. Their view, to the extent they have one, is that the schools are remiss in communicating and explaining test results to parents. 34 3.21 Interestingly, whatever their views of student evaluation, both ministry officials and school directors believe that test results have important consequences. As shown in Figure 3.7, ministry officials believe test results influence their provincial budgets, federal transfers, and public opinion. Figure 3.7: The Province's Evaluation Results Strongly Affect: 60% % 53% 44% 40% 39% 37% 30% 20% 100/ 0% community approval increases in the political consensus in obtaining national provincial budget for the province complementary funds educabon Source: Q6.9, Technical Document 3.22 In-Service Teacher Training. Implementation of the reform included large amounts of training of school staff provided by the national ministry, the provincial ministry, and the school itself. With the exception of school level training financed by the Social Education Plan, school directors report themselves as having very small influence on the type of training received by their teachers. When asked about the value of the training received by staff, directors were highly positive as to the value of the school-specific training financed under the Social Education Plan and highly skeptical at to the value of the training offered by the national and provincial education ministries. Among education ministry officials, 38 percent state that training provided by the national ministry was poor or ineffective (Q6.10). According to ministry officials, many national training courses didn't reflect provincial or school needs, and there was often a mismatch between who was selected to receive the training and who needed it most. 3.23 Federal Role and Strategy. The Federal implementation strategy was in many respects successful. According to provincial ministers, the Council succeeded in creating consensus. The Federal education minister did not use the council to merely inform the provinces of Federal actions and provincial responsibilities. The Social Education Plan was one product of this consensus. The Federal Government, also, delivered technical and financial assistance to the provinces and had some success in improving provincial administrative capacity, training teachers in new curriculum, and renovating schools. No independent assessment of this technical and financial assistance exists, but provincial officials suggest that much of the teacher training was supply driven and not of great 35 benefit. The Federal Ministry was perhaps least successful in communicating the reform to teachers and parents, neither of whom shared the commitment felt by the policymakers. Observers of the implementation of the Reform generally agree in interviews that the Federal Ministry's implementation strategy was not the result of a systematic effort to operationalize the reform by drilling down from the objectives to ascertain what interventions were required for change to take place. Rather the implicit strategy as described here appears to many observers as being ad hoc in nature and may reflect some confusion between levels of government about who is to take the lead in which area or how to work together collaboratively (Table 3.3). Table 3.3: The View of Provincial Education Ministry Officials: Federal Education Responsibilities are Limited Functions The Provincial Government The Federal Government is Most Responsible for is Most Responsible for Student Admissions NA NA Assignment of Teachers to Schools 76.3 NA Curriculum 39.0 47.5 Appointment of Directors 88.1 NA Teachers' Pre-Service Training 33.9 49.2 Teachers' In-Service Training 35.6 45.8 Organization of Work Day 59.3 NA Teacher Salary 76.3 NA Teacher Evaluation NA NA Student Evaluation NA 42.4 Student Promotion 55.9 NA Evaluation of Directors 54.2 NA Administrative Organization 74.6 NA Discipline 27.1 NA Pedagogic Orientation NA NA Teacher Promotion 62.7 NA NA: Fewer than 20 percent ofparticipants believe this level of government is mainly responsible for thisfunction. Source: Q5.1, Technical Document 36 3.24 Argentina's education reform responded to a genuine national consensus on the need for change. Although the reform was legislated nationally, there appears to be less consensus that the Federal Government was the appropriate agent to assume responsibility for implementing that change. The 1998 IIPE survey of educators reveals a "crisis of authority" that calls into question the legitimate role of the Federal Government in education. As shown in Table 3.3 provincial ministry officials believe the Federal ministry has a more important role than the province only in curriculum design, teacher training, and student testing. Even though there is consensus around the need for change, the strong Federal role in implementation is viewed as an imposition on the provinces and the schools. The fact that decentralization has resulted in a national ministry "without schools" contributes to the perception by educators that the policies and programs initiated at the Federal level don't reflect the real world of the school, and so are the technicians of the Federal ministry. 3.25 Our survey results provide further evidence that the Federal Government does not "listen to" and "tailor" its programs to meet the needs of its clients, the provincial ministries. As shown in Figure 3.8, forty percent of provincial ministry officials state that Federal ministry plans and policies "rarely or never" reflect or respond to the specific educational needs of the province. Figure 3.8: One Size Fits All: Percent Responding Federal Policy Does Not Match Provincial Needs 60% T- 55% 50% ~ ~ ~ ~~~~~~5 50%l 46% 40% ~40% 30% 23% 20% - 10% Argentina Catamarca Cordoba Salta Source: Q2.5, Technical Document 3.26 Financial Assistance. As stated earlier, part of the Federal Government's implementation strategy was to provide additional financing tied to the implementation of the Reform. The Social Education Plan (Plan Social Educativo) was created in 1993 to directly address inequities in the education system by targeting resources to schools serving underprivileged children. The Federal Education Pact (Pacto Federal Educativo) was created in 1995 to help the provinces implement the Reform, with emphasis on financing for infrastructure (83% of the total) and equipment. About $400 million was 37 allocated via the Federal Pact between 1995 and 1998. Nationally, total federal education transfers were only 3.4 percent of provincial education spending in 1998. As shown in Figure 3.9, in the three provinces studied this percentage varied from 3.3 percent in Catamarca to 4.5 percent in Salta. 3.27 In transferring financial resources to the provinces, the Federal Government had several options. It could (a) allocate the resources using a formula, leaving the provinces to spend the monies following agreed upon criteria; (b) allocate the resources using a formula, followed by negotiation with the provinces on the precise uses of those monies; or (c) allocate the resources on the basis of requests received from provinces and schools and Federal ministry judgements as to how its financing could leverage the largest gains in Reform implementation. The last was chosen. This mechanism is the least transparent and the most susceptible to possible political influence. We estimated the implicit distribution formula using observations on actual transfers in the Social Education Plan and the Federal Pact across all provinces in the period 1995-1998. Three likely determinants of transfer amounts were included in the regressions: (a) the poverty rate as measured by the provincial index of unmet needs (NBI), since improving equity was one clear goal of the Reform and of the transfers; (b) a variable indicating whether or not the provincial governor was of the same political party as the national executive; and (c) the fiscal capacity of the province as measured by total per capita public sector spending net of these Federal education transfers, as a proxy for the financial need of the province. Figure 3.9: Education Transfers as a Share of Education Expenditures (m. .0 33 3.41 CL 10 A4 0 .5 4.0 193_ 19 3.28~ ~ ~~3. 2he reut ftergesinaayi r rpregnTcncagouet is (D 2.014 a.10 0)~~~~~~~~~. 0.0~~~~~~~. 00Catamarca Cordoba Salta National m 1993 1998 3.28 The results of the regression analysis are reported in Technical Document, first for a pooled data set consisting of the 24 provinces over four years. The results show that politics did not play an important role in determining Federal education transfers and overall, only fiscal capacity is a significant determinant those transfers, and its sign is positive. One possible interpretation of this result is that Federal officials allocate more transfers to fiscally rich provinces because those provinces can more readily provide the counterpart financing and, thus, provide greater leverage for Federal monies. In addition, 38 fiscal capacity may be correlated with the capacity of the province to execute the projects receiving Federal assistance. 3.29 Communication. One means by which the Federal Government tries to lead educators down the road of reform is by communicating the goals of the reform and laying out clearly the sequence of actions required to accomplish those goals. Quite clearly efforts have been made towards these ends, including the publication and dissemination of a monthly education magazine, Zona Educativa. However, in the view of educators and stakeholders these efforts have failed, not because they disagree with the goals and programs but because they do not have a feeling of having participated in creating the solutions to the problems and because they view a lack of alignment between the goals and the resources available to accomplish them. One-third of all school directors state their most serious problem is an inadequate budget (Q2.5D). 3.30 The 1993 Federal Education Law mandated the creation of a national student assessment system to monitor the performance of the country's educational system, to provide feedback to the provinces on their performance, and to identify educationally disadvantaged students and evaluate the impact of Federal compensatory programs on them. Since 1993 the mathematics and language knowledge of students in the final year of primary (EGB2) and lower secondary (EGB3) has been tested. The province-specific results are provided to each province and the school-specific results are posted on the Ministry's website. In addition, school inspectors and directors have been trained in how to interpret and use the results. Despite these efforts, parents complain they do not know the results for their own schools, and directors state that the sample test results are an insufficient basis to guide school policy. Still, the national student assessment system can be rightfully viewed as one of the real successes of the Reform in terms of affecting and motivating behaviors. Over half of all school directors in the provinces studied report that they have undertaken new actions as a consequence of the evaluation results (Q4.14A). C. PROVINCIAL CAPACITY 3.31 While the Federal Government and the Federal Council of Ministers have overall responsibility for implementing the Reform, it is the provincial governments which have the day to day responsibility for carrying out the actions required to accomplish the Reform's goals and objectives. Indeed, given the very limited authority of both the Federal Ministry of Education and the schools themselves, the capacity of the provincial education ministries themselves is critically important to implementing the Reform. In assessing this capacity, we follow the framework laid out in Chapter 2. 3.32 Leadership. Provincial education leadership lies first and foremost with the governor, who guides education policy through his selection of education minister and other political appointments and who plays a key role in determining the level of financial resources available to education. An articulate minister can help construct and guide the government's vision for the future, but if the governor does not view this vision 39 as his own it is likely to die with the next change of ministers, contributing to cynicism among educators. 3.33 The Education Reform offers one vision that governors and provincial education ministers can adopt as their own. It has, also, raised the visibility of education as a political issue and has, perhaps, increased the political benefits of improving education. As shown earlier, most provinces have increased coverage at the lower secondary level and, also, increased provincial funding for education. However, few provinces have constructed a truly provincial vision building on the foundation of the Reform. 3.34 In the provinces surveyed for this study, the dominant provincial vision of education is one of an efficient bureaucracy without the political patronage of the past, with improved managerial controls to avoid abuses of medical leave and to eliminate ghost teachers, and with streamlined ministry management. This vision matches well the fiscal crises several provinces have confronted in recent years. However, the provinces have, also, had problems of poverty, high youth unemployment, and low economic growth, and one does not in general find visions about the role of improving access to and quality of schooling in resolving these problems. 3.35 Commitment. Since participation helps create commitment, one means of measuring commitment is assessing ministry officials' own perceptions of their involvement in the Reform. Overall, 60 percent of education ministry officials report the province was highly involved in setting priorities, and 54 percent report it was highly involved in determining the pace of implementation. 3.36 Authority. As was shown in Table 3.3, provincial education ministry officials believe they have the authority to implement most aspects of the education reform. The provincial ministry sets pay, recruits teachers and directors, evaluates directors, and determines its own administrative organization. It is only in the areas of curriculum design, teacher training, and student assessment that provincial officials believe the Federal government has greater responsibility. Ironically, two of these areas of shared responsibility-curriculum change and teacher in-service training-are the highest priority problems, according to provincial ministry officials. The ambiguity as to the role of the province in teacher training may partly explain the limited success to date in reforming teacher training institutions under the reform. 3.37 Incentives. Incentives for staff performance are relatively weak in the provincial education ministries. For example, only 33 percent of ministry officials report that "employees who appear to be competent and hard working are treated much better than employees who do not appear to be competent and hard working (Q62)." And 28 percent of officials report that "in the past three years, staff in my organization have been recognized or rewarded for taking prompt action on citizen requests or complaints or providing outstanding service (Q67)". 40 3.38 On the other hand, provincial education officials believe that there are relative strong financial incentives to implement the reform, as shown in Figure 3.10 below. Figure 3.10: Financial Consequences of Implementing the Reform 80% j 700/o 67% 690/o 50%~ 400/o 30%- 100/0 obtaininq obtaininq national inaeases in the conrmnity approval intemational credits comrlementarv provincial budqet for funds education Source: Q6.4, Technical Document 3.39 Accountability. In response to the question (Q6. 17) of whether adequate accountability mechanisms exist, provincial ministry officials provide a clear response. They view their principal client as being the school and its students and teachers. Focus group discussions confirm this view. However, these discussions, also, indicate that officials information flows mainly in one direction-from top to bottom. No mechanisms, aside from the visits of school inspectors, exist to systematically inform officials about the needs and problems of their clients, i.e., how best to serve them. Ministries rarely, if ever, survey their clients to obtain unfiltered feedback on their views of the services offered. Perhaps for this reason, schools see provincial education initiatives as failing to meet their needs and, in fact, less closely aligned to their needs than the initiatives of the more distant Federal ministry. 3.40 While there are no mechanisms in place for accountability to the school client, there are reporting and control mechanisms to ensure accountability to the principal financier of education-the provincial government. These mechanisms emphasize expenditure control and, hence, are input rather than output oriented. If the ministry is doing a bad job of providing service to the schools of the province, there is almost no way the education minister, the governor, or the legislature would know. 3.41 Resources. While educators rarely think financial resources are adequate, education ministry officials do not believe it is the most important problem. Only 15 percent of provincial education officials report that an inadequate budget is the cause of the major educational problems facing the province (Q3. l). 41 D. CAPACITY OF THE SCHOOL 3.42 While the provincial government has a near monopoly on educational decision making in most provinces, the changes in the school and classroom required to successfully change the curriculum and organization of teaching are the responsibility of the school staff. Hence, it is appropriate to ask what is the capacity of the school to implement change. The assessment given here draws from the survey results; interviews and focus groups held with school inspectors, directors, and parent groups; and earlier surveys of schools, teachers and parents carried out in Argentina. 3.43 Leadership. The school director has little authority, apart from the power to persuade colleagues to collaborate, and even less preparation to take on the mantel of leadership. The authority of the director is reduced by the nature of the appointment, which is more often a reward for longevity rather than demonstrated competence. This environment does not preclude a director from exercising leadership and working with the school community to develop a shared vision, but it does make it unlikely. 3.44 Is there clarity in the identification of problems or the specification of goals, especially as concerns outcomes (as opposed to inputs)? School directors report that that national testing system has given them for the first time objective evidence on their performance, and that evidence has been useful in setting overall school goals in the SDP. However, test results have not been accompanied with the kind of school inspection or evaluation which would guide the school in defining its problems. In addition, as noted earlier, there is a lack of clarity for many teachers and directors as to the specific outcome goals of the reform, aside from expanding coverage and reorganizing instruction. The result has been that most schools state their goals in an ambiguous fashion that fails to provide a focus for the actions of the school community. 3.45 Commitment. The key actor in the school is the teacher, and as noted earlier teachers often do not understand and are not committed to the goals of the reform and their key role in accomplishing those goals. This is true despite the attempts by the Federal and provincial education ministries to communicate the reform and its goals to teachers. Furthermore, Argentina's experience with engaging teachers in reform is not that different from the experience of other countries (Kirst, 1970; Sarason, 1990; Elmore, 1996). A major explanatory factor for this outcome in Argentina's case may be the perception by teachers that they did not participate in defining the reform and that the reform has been imposed on them (IIPE, 1998). 3.46 Authority. As noted earlier, the schools have very little autonomy of decision making power in Argentine public education; almost all important decisions are centralized at the level of the provincial ministry. Schools do not select their own personnel, do not determine or execute their own budgets, have relatively little voice regarding the curriculum, and have almost no say regarding the training of their staff. On the other hand, as shown in Figure 3.11 below, the role of the school and, especially, 42 teachers is critical in preparing its SDP, and implementation of the new curriculum requires that teachers change what they do in the classroom. Focus groups held with school directors and inspectors and other, anecdotal evidence suggests that the SDP is not being implemented well. While almost two-thirds of school directors surveyed report that the principal objective of the SDP is to improve the quality of education (Q2. 1), there is a disconnect between this objective and the actions proposed, perhaps in part due to the fact that the school must propose actions that have some possibility of being financed by the provincial or national authorities (IIPE, 2000). Figure 3.11: Teachers Dominate the Preparation of the School Development Project 120%- __ 100% 98% 80% I 60% 40% - 29%2 21% 19%. 19%~~~~~0 20%~~~~~F7 . 10% ;.H 0% community inspectors students parents school teachers members Source: Q2.3, Technical Document 3.47 More importantly, as shown in Fig. 3.12 below, the schools have very little say over the training of their teachers, either in terms of identifying teachers to be trained or in selected the subject matter of the training. School directors report that by far the most valuable training is that which they requested and was delivered in their own school, with financing from the Social Fund. The supply driven nature of teacher training does not allow schools to match training to their SDP objectives and doesn't allow the director to match an individual's training to that teacher's career development goals. 43 Figure 3.12: The School Has Little Say in Teacher Training 90% - 80% 70% 60%H 50% n n 4 E No Influence 30% 2 A% 640A 40%/0 10 91 2 ,_ High Influence 10%- 10% - , 0% ------ _ r-_ _--__ I c ' O2 a 2 in C2 U) (D 2 z~ Source: Q 5.4A, Technical Document 3.48 Incentives. There are few positive financial incentives in the public school. The systems of teacher and director evaluation are perfunctory, with 80 percent of all teachers rated as having "excellent" performance (Q4.5, Technical Document). The only penalty associated with poor performance is possibly a reduced probability of promotion. Most teachers don't even receive feedback on how to improve their teaching performance. Neither the performance of directors nor teachers is linked to pay or job stability. Few school directors report that their school's score on the national assessment test has important consequences (Q4.1 4A). 3.49 In addition to incentives for performance, one can ask whether there are incentives for teachers to participate in the preparation of the SDP, to participate in training programs, to engage parents in the schooling of their children, or to work with colleagues to improve teaching practices. All of these activities require additional effort or hours worked by the teacher without additional compensation. Training programs may lead to better credentials and, eventually, higher pay, but participation in other activities offers only non-pecuniary benefits. 3.50 Accountability. The school has two clients-the parents of its students and the provincial ministry which finances its operations. As noted earlier, parents are only marginally involved in the schools. Very few of them know their school's performance on that national assessment test. They by and large do not participate in the design and preparation of the school's SDP, and the school cooperatives are restricted solely to raising money. If dissatisfied with the school, parents can move their children (Q5.8A, Technical Document), but very few do so, and there are no financial consequences for the school. In addition, the process of school selection is not transparent (Fiszbein, 1999). Despite the evidence student learning is low relative to the OECD countries Argentina aspires to join, over 90 percent of parents express strong satisfaction with their own school, teachers, and director (IIEP, 2000). 44 3.51 Accountability to the provincial ministry is equally weak. School inspectors seldom have the time and resources to visit schools sufficiently to actually know the problems of the school, although the test scores available on some schools through the national assessment system provide significantly better information than they had previously. There are almost no consequences of poor performance. It is rare to remove a school director for this reason. 3.52 Resources. School directors report that the financing they receive from the provincial and national governments for special school projects is by and large unrelated to their needs and is unpredictable in terms of both timing and amount. However, the schools do have some capacity to seek external financing from the private sector (companies and foundations), and two-thirds of school directors report that it would be possible to obtain additional financing from these sources (Q2.1 0, 2.1 1 A). 3.53 In summary, the school has few of the characteristics that would contribute to success in implementing school and classroom reforms. 45 4. CHAPTER HEALTH: PROVINCIAL EXPERIENCE IN IMPLEMENTING A PILOT REFORM IN HOSPITAL MANAGEMENT CHAPTER SUMMARY In one of a series of executive decrees aimed at improving efficiency in the health care system, the Federal Government in 1993 issued Decreto 578/93 which reaffirmed the right of the provinces to give their public hospitals management autonomy and allowedfor the automatic cost recovery by public hospitals from health insurers through the national social security fund.. Subsequently, the Government created a project, with World Bank support, to provide technical assistance to several provinces to implement the decree in selected public hospitals. This chapter looks at the background leading up to the reform, assesses Federal and provincial roles in implementing the reform, presents some of the results, and ends with a discussion oJ how this Dilot reform mieht be taken to scale. A. REFORM OF HOSPITAL MANAGEMENT 4.1 Early in the 1990s, the Federal Government issued several executive decrees to introduce greater competition and efficiency in the health care system. The emphasis of these decrees was on the reform of health insurance (Obras Sociales) in Argentina to encourage competition and choice among health care providers. This reform has been subjected to extensive review and assessment which will not be repeated here.'6 This study focuses, instead, on the important but largely ignored Federal effort to encourage provincial governments to convert their traditional public hospitals into self-managed organizations.17 The potential importance of this reform effort is demonstrated by the fact that public hospitals represent 40 percent of total bed capacity in Argentina and consume 80 percent of the total public health budget. 4.2 Context of the Reform. Argentina's health indicators are good when compared to other countries in the Latin America region but worse than those of other middle income countries and lower than would be expected given the country's level of income and education. For example, Chile, with lower per capita income than Argentina, had in 1995 an infant mortality rate of 12 per 1000 live births, while Argentina had a rate of 22 per 16 See, for example, Montoya and Colina (nd), World Bank (1997). 17 This reform was preceded by at least two somewhat similar reforms. In the 1960s, the Province of Buenos Aires experimented with Reform Hospitals, and in 1967 a Federal law gave hospitals (SAMIC) the legal basis required for autonomous decision making. 46 1000 live births. Also, in 1995, Chile had a maternal mortality rate of 6.5 per 10,000 live births, compared to Argentina's 14 per 10,000 live births. Table 4.1: Correlation Between Per Capita Household Income and Years of Life Lost per 1,000 Population in Argentina Disease Correlation Diarrhea -.46 Tetanus -.46 Malnutrition -.45 Abortion -.44 Syphilis -.42 Chagas' Disease -.38 Tuberculosis -.36 Congenital Heart Disease -.32 Cervix Uteri Cancer -.32 Source: World Bank (2000), Argentina: Poor People in a Rich Country 4.3 Within Argentina the poor die earlier than they should. There are strong negative correlations between per capita household income and health indicators such as the infant mortality rate (r=-0.62), the maternal mortality rate (r=-0.48), and the proportion of low birth weight babies (r=-0.58).18 As shown in Table 4.1, there is, also, a strong negative correlation between per capita household income and the years of life lost due to communicable and preventable diseases. Table 4.2: Quality of Maternal Care by Household Income per Capita |_____________ _ __ Per Capita Household Income Quintiles _ __ _ _ _ _ _ _ _ _I__ _ 1 2 3 4 5 Prenatal care delivered by non 8.3 3.3 2.1 0.4 0.1 doctor Delivery by non doctor 28.5 17.9 10.2 8.2 0.1 Cesarean section 21.8 33.7 37.7 41.5 42.1 No post partum follow up 31.6 17.2 6.5 2.5 2.3 Source: Social Development Household Survey 4.4 While Argentina enjoys an ample supply of health care (367 inhabitants per doctor and 209 inhabitants per hospital bed), access to health care is highly correlated with income. Also, there is evidence that the quality of care also varies with income. For example, using prenatal care as a tracer for the quality of care, we find a direct negative association between per capita income of the household and the probability of having had less then five prenatal exams. Most of the poor have access to public hospitals, but the care they receive differs significantly from that received by higher income groups. As shown in Table 4.2, over 25 percent of pregnant women coming from the lowest income quintile are delivered by non-doctors, compared to one tenth of one percent of women from the highest income quintile. 4.5 As shown in Table 4.3, part of the explanation for differences in health care across income groups lies with the differences in insurance coverage. While the public health service covers the entire population, 91 percent of the highest income quintile are covered by health insurance and thus have the option of choosing private health care providers. The lowest income quintile, on the other hand, must by and large depend on Is See World Bank (1999), Poor People in a Rich Country: A Poverty Report for Argentina, Report No. 1992AR. 47 the public hospital for care; only one-third of the population in the lowest income quintile are covered by some form of health care insurance. I Table 4.3: Health Insurance Coverage, Public Health Expenditures, and Private Health Expenditures by Income Quintile Income Percent of Households Share of Private Health Quintile Lacking Health Provincial Health Expenditures as Insurance Expenditures Share of Income 1" 65.8 26.6 5.2 2"d 38.4 19.8 7.3 3rd 26.3 22.0 8.2 4th 16.3 18.5 9.4 5th 9.3 13.1 9.5 Total/ 35.3 100.0 8.6 A verage __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Source: Adapted from 1997 Social Development Household Survey and 1996/97 National Household Expenditure Survey as reported in World Bank (2000) 4.6 Given the dependence by the poor on public health care, the management of public hospitals directly affects their welfare. The fact that public hospitals have traditionally offered care free of charge results in insured individuals using public hospitals to avoid co-payment, especially for expensive services. Approximately 30 percent of all public health services are provided to the insured free of charge, resulting in a significant cross-subsidy to the private sector. As shown in Table 4.3, almost two- thirds of the lowest income quintile lack health insurance coverage, yet they receive only 26 percent of total provincial public health expenditures. 4.7 The traditionally managed public hospital does not charge for the use of services, has no incentive to ration services to the insured in order to improve services to the poor, and has no incentives to provide services of high quality or at low cost. Public hospital budgets are determined on the basis of historical patterns and bear no relationship to the quantity or quality of services offered, or the capacity of patients to pay. On the other hand, budget constraints prevent public hospitals from providing free medicines to the poor, resulting in the poorest quintile of households spending over 5 percent of total household expenditures on health care. 19 For detailed data on the use of public hospitals see the Siempro web site -www.siempro.gov.ar 48 Figure 4.1: Relation Between Population NBI and Infant Mortality Rate 2] Salta o12 Caaac 10 5 O 1 0 2 0 3 0 4 10 6- 2 Cordoba 0 5 10 15 20 25 30 35 40 45 Populabon NBI 4.8 In addition to differences across income groups, there are, also, large variations in health indicators across the provinces. As shown in Figure 4.1, there is a strong correlation between a province's poverty rate, as measured by the percent of the population with unmet basic needs (NBI), and health indicators such as infant mortality. As shown, given its NBI, Salta's infant mortality rate falls on the regression line, but Catamarca's lies above, and Cordoba's lies below, suggesting that Salta and Cordoba are more effective than Catamarca in reducing infant mortality. This suggestion is further reinforced in Figure 4.2, which plots infant mortality against per capita public expenditures on health. Here we find both Cordoba and Salta have lower infant mortality than expected given their expenditures, while Catamarca has higher than expected infant mortality. It is important to note that these Figures use data from the mid-1990s and do not reflect the progress Catamarca has made since that time in reducing infant mortality. Figure 4.2: Relation Between Infant Mortality Rate and Public Health Expenditure Per Capita 14 l 12 Catamarra 10- a) * 6 1 National Salta 4 2 Cordoba 0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 Total Public Expenditure per Capita 49 4.9 Objectives and Actions. As noted above, the National Government has embarked on several major national level reforms of the health insurance system to improve its efficiency and equity. In addition, several provinces have, with World Bank assistance, initiated the implementation of gradual health reforms in their jurisdictions. Although the strategies and emphases, as well as the implementation stages, vary from one province to another, most of these reform efforts aim at: (i) expanding insurance coverage for the uninsured population (with a basic package of health benefits focused, initially, in regions within the province with lower income levels); (ii) improving the performance and management of the major public hospitals (including increasing cost recovery from third- party payers and introducing new performance-based incentives); (iii) strengthening the primary health care network; and (iv) bringing the provincial insurance carrier to financial equilibrium. 4.10 The implementation of provincial health sector reforms to improve the quality and quantity of services provided to the poor thus requires that hospitals have the capacity to manage themselves. Decree 578/93 reaffirmed the right of provinces to give public hospitals management autonomy, to charge fees of insured patients, and, most importantly, allowed for the automatic cost-recovery by public hospitals from health insurers through the national social security fund. 4.11 Implementation Strategy. In addition to giving provinces the legal authority to grant their public hospitals management autonomy, two other policies were adopted to facilitate implementation of the Decree. First, the Federal Government health ministry provided technical assistance and civil works investments to selected provinces and hospitals (in Mendoza, Province of Buenos Aries and City of Buenos Aires) under a World Bank financed project called the Provincial Health Sector Reform Project, or PRESSAL (Proyecto de Reforma del Sector Salud en las Provincias). Second, other provinces were authorized to borrow from the Bank for the purpose of fiscal management along with social sector improvements, with measures that encourage hospital autonomy. Salta was one of the provinces joining the Second Provincial Reform Loan in 1998, while Cordoba and Catamarca have now finished negotiating reform packages with the Bank.20 B. IMPLEMENTATION OF THE REFORM 4.12 The provincial governments had the freedom to choose or not to send a letter of intent to the Federal Government stating that they would commit themselves to converting the hospitals under their jurisdiction to self-managed hospitals as described in Decree 578/93 (Article 18). The incentive for sending such a letter of intent was very strong, as the Decree required the Social Security system (ANSSAL) to automatically pay for services provided by provincial public hospitals only in those cases where the province indicated its agreement with the decree. However, signing a letter of intent did 20 All three provinces-Catamarca, Cordoba, and Salta-were among those participating in the first round of transfer of national hospitals to the provinces in 1978; as noted by Bisang and Cetrangolo (1997), this transfer of responsibility was not accompanied by a transfer of financing. 50 not obligate the province to introduce reforms according to any specific timeline. In response to these incentives, the provincial governments affiliated with the national party in power-the Justicialistas-immediately joined the reform, and almost all other provincial governrents also joined the reform, once it became clear that doing so did not necessarily require that the decree be implemented. 4.13 Provincial Implementation Experience. Hospital management entails a wide variety of functions each of which may be centralized in the provincial health ministry or decentralized to the hospital. These functions include governance, strategic planning, administration, purchase of inputs, investments, financial management, personnel management, and business management. 51 Table 4.4: The Characteristics of Public Hospitals by Degree of Autonomy Function Centralized Decentralized Governance No board of directors; hospital director Board of directors is named by the community; hospital named by the Ministry of Health director named by the board. Strategic Planning If a strategic plan exists, the Ministry has The strategic plan is developed by hospial staff and the designed it. board of directors; no Ministry approval required.. Administrative Manual Hospital rules dictated by the Ministry. Hospital managers determine their own rules; Ministry approval not required. Input Purchases The Ministry makes all majorpurchases and The board of directors and hospital staff make all distributes to hospitals. purchases. Investment The Ministry makes all investment decisions. Under the control of the board of directors and hospital management. Financial Management All funds are public. The budget and its Budget is developed and executed by the hospital on the allocation within the hospital is determined basis of estimated services provided to public and private by the Ministry. sources of finance. Personnel Hospital staff fall under the civil service of Board of directors makes all decisions regarding pay and Management the province staffing, subject to private sector regulations Sales Management The hospital doesn't contract to sell services. The hospital freely contracts with public and private I________________ I_______________________________ finance sources and with other hospitals. Source: Technical Document 4.14 Table 4.4 describes each of these functions as they might be found in (a) a highly centralized system and (b) a highly decentralized system. While Argentina's hospital autonomy decree did not specify the degree of decentralization to be expected by each of the functions listed in the table, over time movement was expected in the direction of decentralization as defined. As described in the Technical Document, the actual degree of decentralization for each of these functions was assessed for the few of the country's major public hospitals where the PRESSAL program had provided assistance.21 In addition, the assessments across functions were weighted to develop a one hundred point scale. The results are reported in Table 4.5 for selected hospitals. Mendoza has progressed furthest among the provinces in terms of hospital management autonomy. This is attributable in large part to the fact that the Mendoza hospital management reform predated the Federal reform, with the passage of Provincial Law 6015 in 1993. That law gave the hospital director substantial authority to purchase, contract, and charge for services delivered. Table 4.5: Hospi talAutonomy in the Pro inces, 1998-2000 Province Hospital Legal I Govern- Strategi Admin. Input Invest- Fin. Personnel Sales Overall Basis ance c Plan Purchases ment Moi Mngt. .MnRt Catamarca SanJuan 0 0 0 3 0 0 2 5 2 14 Batista _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Cordoba Matemidad 0 0 2 3 4 6 5 5 2 27 Salta Onatavia 9 6 1 1 8 0 12 15 8 60 Mendoza Lagomaggiore 9 6 8 5 10 9 8 10 2 67 ICentral 9 6 s 4 8 10 l 8 10 8 71 Source: Technical Document 4.15 For the provinces reviewed in this study, Salta has advanced most in the direction of hospital autonomy. Public hospitals have been given their own legal status; assets and personnel have been transferred to the hospital corporation; and the board of directors is elected. Significantly, hospital autonomy in Salta has been only one element in an overall health reform that goes far beyond Federal Decree 578/93 and that has been 21 In addition, the Federal Ministry of Health carried out a 1999 evaluation of the implementation of Decree 578/93 in 287 hospitals throughout the country. The results showed a high degree of formal, legal compliance. 52 implemented gradually since the current Governor took office in 1995. Other elements of the Salta reform include reform of the provincial health insurance plan and separation of supply from finance through the creation of a public agent to purchase services on behalf of the population not covered by public or private health insurance. 4.16 Up until 1999, Catamarca and Cordoba had advanced far less in implementing hospital autonomy. Catamarca passed a law of hospital autonomy but failed to write the implementing regulations. Cordoba, which like Catamarca was in political opposition to the Federal Government over the past five years, was one of only two provinces to fail to even sign a letter of intent to implement Decree 578/93. It is important to note that there have been other health reforms implemented by provinces that are not directly related to hospital autonomy, but to other health fields. That is the case of Catamarca that initiated an aggressive promotional and preventive sanitary policy. 4.17 Stakeholders and Their Views. The implementation of the hospital autonomy reform includes several actors with differing interests and objectives. Notably absent from this discussion of important stakeholders is the patient or client, who has almost no power, either individually or collectively. It's difficult to find provincial consumer associations which purport to represent the interests of patients. 4.18 Perhaps the single most important actor is the governor of the province, who sets the political priorities of his government and who chooses ministers of health and economy who will take actions consistent with those priorities. A governor can be a proactive reformer, be hostile to Federal reform efforts, or somewhere in between. The political calculus which lies behind the governors' decisions and actions is complex and, judging by the few governors who have taken health reform seriously, is generally found to produce net political costs. 4.19 While government ministers are charged with implementing the governor's program, they also have their own interests which may oppose reform. For example, the ability of the health minister to directly name the hospital director and to respond to requests for supplies or personnel is a source of political power that he may not willingly cede. Similarly, an economy minister charged with reducing fiscal deficits may have a strong interest in centralizing financial responsibilities in order to control spending. In some provinces, economy ministers have centralized the collection of fees for services public hospitals provide to patients with health insurance (obras sociales and pre-pagas), thereby greatly reducing the scope for hospital autonomy. Based on interviews with provincial officials, those responsible for provincial finances have viewed hospital autonomy reform as a threat and not as an instrument for controlling spending and improving efficiency. 4.20 Health ministry staff may also oppose hospital autonomy and have various means to discourage its implementation. For one thing, many ministry staff do the kind of work-contracting, purchasing, personnel management, etc.-which would face greatly reduced demand with hospital autonomy. For another, the combination of low public sector salaries with high power to direct public spending for hospital inputs provides fruitful conditions for corruption. 53 4.21 Two unions are important stakeholders in the health sector. The union of provincial employees includes almost all non-physician staff. This union sees hospital autonomy as posing numerous threats: (1) loss of negotiating power as a result of the central ministry losing its most important functions; (2) possible increase of temporary over permanent staff; (3) risk of layoffs as hospitals seek greater efficiency; and (4) loss of membership if public employees are transferred to hospitals with private legal status. On the other hand, a few health worker unions, like FATSA in Salta, which represent both public and private hospital staff, have supported autonomy because of the prospect of more members and higher pay in privately managed hospitals. Also, in Salta, the provincial government gained the support of the health union for contracting out services by granting long-term contracts to new firms that would continue to employ unionized hospital staff. 4.22 Another important union is the physicians' association (Colegio, Circulo or Sindicato), which has complicated interests since most members have private practices as well as full-time staff positions in public hospitals. Among the fears of this union are: (1) possible loss of labor stability; (2) new salary regimes which may reduce pay for some physicians; (3) requirements that physicians receiving full-time pay also work full time; and (4) decreased demand for their private practices, as public hospitals begin charging and reducing what insurers can pay for private care. In addition, in those provinces (e.g., Catamarca) where the physician's union has played an intermediary role between the physician and the insurer, demand for this function may diminish as hospitals opt to charge insurers directly. 4.23 Another key stakeholder is the management of the public hospital. Based on interviews with hospital officials, here one can find the unabashed supporters of reform, who chafe at the constraints and inefficiencies in the current system and who want a level of authority consistent with what they perceive as their responsibility. On the other hand, there are also hospital managers who have gained their position through political loyalty rather than technical skills and who may fear the unknown risks of a more open and competitive environment. 4.24 Federal Role. The decentralization of public hospitals from the Central Government to the Provinces, combined with financing through unconditional revenue sharing (co-participacion) rather than sector-specific grants, has greatly reduced the role of the Federal Government in hospital care and, also, reduced its capacity to bring about hospital reforms. The Federal Minister of Health still retains considerable power to influence the actions of the provincial ministers through his power to convene the Federal Council of Health Ministers and to use that Council to advocate a reform agenda. There is little evidence, however, that the Council was used in this way in the period of time studied here. Participants in Council meetings describe them as used mainly for the Federal Ministry to inform the provinces of Federal actions and not to stimulate discussion and reach consensus. In addition, the Federal Ministry has other instruments to influence provincial ministries, including the distribution of drugs, vaccines, etc., associated with specific national health programs; funding from international credits; and hospital regulations. 54 4.25 Some observers allege that the hospital autonomy reform, which enjoyed early support at the Federal level as part of the overall health reform to bring about greater competition and efficiency, was subsequently "orphaned" after a change of Federal health minister. As a result, the only influence of the ministry was its ability to provide credits for hospital infrastructure and free technical assistance, which was productive only when provincial governments already had reform agendas. The Bank-financed PRESSAL program, which provided this assistance in several provinces, financed management studies and training, in addition to physical infrastructure improvements. C. PROVINCIAL CAPACITY 4.26 The capacity of provincial health ministries to manage the planning and delivery of health services clearly affects the implementation and sustainability of reforms, including hospital autonomy. As described earlier, to assess capacity we analyzed the answers given by health ministry officials to the Public Officials Survey, and, also, we interviewed a wide range of stakeholders, including hospital administrative staff. In addition, the experience of the Federal Health Ministry in working with provincial ministries in implementation of the PRESSAL project provides a rich experiential data base. 4.27 Self-Assessment of Effectiveness. Provincial health ministry officials do not hold their own ministries in high repute. As shown earlier in Figure 2.5, health officials rate the effectiveness of their ministries lower than any other sector surveyed. Only 7 percent of health officials affirm that their ministry "has been highly successful in fulfilling its objectives." On another question (Q27) concerning the expected performance of ministry staff, again only 7 percent report that "in my organization performance standards are met or exceeded very frequently." This is far lower than the 48 percent in the provincial education ministries or the average of 21 percent in all provincial ministries. 4.28 Both corruption and political influence are serious problems according to ministry officials, and these may partly explain their views on organizational effectiveness. Of all respondents, 59 percent view corruption as a significant problem, and only 7 percent of officials affirm that corruption cases are reported to the proper authorities, a percentage considerably lower than that found in other ministries surveyed. Finally, more than half of health ministry respondents say that political interference is frequent and influences hiring and promotion (Q58, Q59). 4.29 Leadership and Commitment. Federal Decree 578/93, while titled a decree of hospital autonomy, in fact provided only a partial vision of the "new public hospital." Indeed, the major focus of the decree was on cost-recovery rather than autonomous management, and it made little if any mention of the various elements required for hospitals to manage themselves. Ambiguity as to the vision resulted in diverse readings or interpretations of the decree and a continued lack of shared understanding as to the meaning of, rationale for, and consequences of hospital autonomy. 4.30 Despite signing letters of intent to implement the decree, few provincial health officials understood the meaning of hospital autonomy. Furthermore, the national 55 Ministry of Health did not undertake a strong communications campaign to educate stakeholders about the reform. Indeed, as noted earlier, almost two-thirds of all survey respondents report that they never discuss health policy and performance with counterparts in the national government (Q50), the highest percentage among the social sector ministries. Poor communication between levels of government is mirrored in the communication within provincial ministries, as measured by policy consensus. Almost half of all health officials report disagreeing with the policies the ministry itself is asked to implement (Q56). 4.31 One result of poor communication is a widespread lack of knowledge about hospital autonomy among key stakeholders. Many fear that autonomy inevitably means public hospitals will compete for "paying" customers and thus exclude the poor, or they will offer lower service quality to non-paying customers. Since Argentine health care professionals are proud of the country's record in providing health care to all, this fear translates into opposition to the hospital reform. The fact that this reform did not take place in the context of a more comprehensive health sector reform that provided health insurance to the poor lends credence to this fear of greater inequity. However, most of the characteristics of autonomous hospitals listed in Table 4.4 can be attained in the absence of cost-recovery and without any significant risk to equity.. 4.32 As noted earlier, merit-based recruitment practices may contribute to the credibility that influences commitment. The picture for provincial health ministries is mixed. Recruitment does not appear to be open. Only 13 percent of survey respondents in health report that "positions like mine are generally advertised to the public (Q10)," and only 1 percent report finding their own job from published advertisement (Q9), the lowest among all provincial ministries. On the other hand, 34 percent report that hires are done "primarily on the basis of merit (Q15)", by far the highest percentage of all provincial ministries. 4.33 Authority. Ever since public hospitals were decentralized in the 1970s, the provinces have had a great deal of autonomy and authority to manage their own health care systems. Decree 578/93 extended and clarified this authority with respect to decisions to create self-governing boards of directors and to impose user fees on the insured. In addition, health staff themselves by and large believe they have adequate authority; 64 percent report they "have the appropriate amount of discretion in the performance of their duties (Q28)." Another indicator of decisionmaking discretion as perceived by health ministry staff is their ability to design their own policies and programs, as opposed to simply following Federal directives. Compared to other social sector ministries, a significantly higher percentage (58%) of health officials report "We adapt Federal Government programs to our particular needs (Q52)." 4.34 Incentives. The health ministry's effectiveness is influenced by the incentives for health staff to align their own actions with ministry goals. One incentive is a fair and frequent staff performance evaluation, with consequences for pay and promotion. Among health ministry staff, almost all staff (94 percent) have had their performance evaluated within the past five years (Q22), by far the highest percentage among the social sector ministries. Also, most (54 percent) report the evaluation as being fair (Q24). On 56 the other hand, 51 percent report that political connections are very important in getting ahead (Q62), and 61 percent report that staff have not been disciplined for poor performance in the past three years (Q64). 4.35 Other incentives were noted earlier. The PRESSAL program offered the "carrot" of physical infrastructure improvements to a few provinces more advanced in hospital self-management although in practice this carrot often preceded reforms and had little real impact. Indeed, some would argue that financing physical infrastructure as part of a health reform project almost automatically puts the focus on procurement and construction and reduces the focus on reform. Expecting to see concrete institutional improvements-particularly in terms of patient care, facility utilization, and financial management-while a hospital is undergoing the disruption of major construction efforts may be unrealistic. In addition, the Provincial Reform Loans supported by the Bank included measures s supportive of health reform, including hospital autonomy but did not include any support for constructions.22 Bank experience shows that such efforts can be effective when they have the genuine support of the government. 4.36 Accountability. The hospital autonomy reform is designed to improve accountability by establishing independent boards of directors and creating appropriate performance incentives. Strengthening the competition for clients also helps ensure accountability to the client. Health ministries also have clients to whom they may be held accountable. One client is the ministry of economy, which is mainly concerned with the financial performance of the sector, and another is the public hospital, which depends on the ministry for numerous services. Accountability to either is weak. The perception of health ministry officials is that overspending is not penalized; 90 percent of respondents say there is no penalty associated with the ministry exceeding its authorized budget (Q43). Similarly, there is no penalty to the ministry associated with slow delivery or low quality of services to hospitals. Indeed, information on ministry performance-either measured or as perceived by their clients-is neither collected nor available. As with other ministries, health officials report that their ministry keeps financial records but it's very difficult to gain access to them (Q3 1). 4.37 Resources. Adequate human and financial resources are critical to an organization's effectiveness in implementing reform. Both insufficient resources and unpredictability of resources-either rapid turnover of key personnel or uncertain budgets-adversely affect effectiveness. There is, if anything, very slow turnover of health staff; more than half have been in the ministry over ten years, and 76 percent expect to be there at least another five years (Q4). On the other hand, the health ministry is similar to other provincial ministries in terms of unpredictable financing. Actual funding typically falls far short of budgeted funds, as reported by health officials (Q36, Q38). 22 In comparison the Bank-financed Health Project in Uruguay provided support only for institutional improvements, and no construction, in a pilot of four public hospitals. The experience in Uruguay also demonstrates that there is considerable potential for realizing gains in effectiveness, quality and efficiency of public health care without costly and disruption construction programs. In the case of a Bank-financed Health Project in Chile, the experience was similar to that of Argentina: that is, even after considerable investment in improving facilities, there was little change in hospital management practices. 57 4.38 Summary. Health ministries are not greatly different from other provincial ministries, in the perception of their staff. They are perceived as being less effective overall. In general, strong leadership for reforn has been lacking, and ministry staff and other stakeholders lack commitment to reform. Ministries have the necessary authority to make decisions, but incentives for performance, either at the individual or organizational level, are weak. Perceptions of corruption and political influence, also, adversely affect incentives for aligning with organizational goals. Financial resources are by and large adequate but are unpredictable and used inefficiently. 58 5. CHAPTER LESSONS LEARNED CHAPTER SUMMARY What lessons have we learnedfrom Argentina's experience in implementing education and health reforms so far? What would we recommend to future social sector reformers at the Federal or provincial levels of government in order to continue to advance the reforms in health and education? What would we recommend to those who wishes to initiate reforms in new areas, such as nutrition, training, unemployment assistance, etc.? To answer these questions, in this chapter we first summarize the findings obtained from the assessment of Argentina's implementation experience. These findings points to a very substantial need to modernize administrative capacities at the provincial level and to address systemic problems of corruption, accountability, and incentives. s. Finally, we discuss the implications oj the lessons learned to date for the implementation offuture reforms. The education and health experience suggests a successful Federal implementation strategy would, among other things, (a) use the council of provincial ministers to bring about provincial consensus; (b) undertake a mass campaign to educate beneficiaries and stakeholders on the nature and consequences of reform; (c) provide technical and financial assistance to provincial governments; and (d) carefully monitor and evaluate provincialprogress. A. SUMMARY OF FINDINGS. 5.1 This study set out to attempt to assess the constraints to the implementation of federally initiated social sector reforms in Argentina and to advance ways in which these reforms can have greater impact on service delivery in the provinces. The raw data for answering this question came from a sample of the experience of implementing two reforms-in education and health-in three provinces-Catamarca, Cordoba, and Salta. The focus of the study was not to evaluate the reforms themselves but, rather, to evaluate their implementation. 5.2 Issues Addressed. To attempt to answer the questions posed above, the study examined [a] the context within which the reforms were introduced and implemented; [b] the formulation and design of the reforms themselves, [c] the Federal Government strategy for implementing the reforms, [d] the views of key stakeholders towards the reforms, and [e] the capacities of provincial government ministries and service delivery 59 units to manage change. Given the critical importance of the provincial ministries, special attention was paid to assessing their implementation capacity. Surveys and interviews were conducted to assess six key characteristics of capacity: [a] leadership and clarity of goals, [b] commitment and ownership, [c] authority, [d] incentives, [e] accountability, and [f] resources. 5.3 Results of Reform. Although the sample of reforms and provinces was small, there was considerable variation in terms of implementing the reforms. The education reform was, in general, more effectively implemented than the health reform, although not all elements of the education reform were equally well implemented. Among the provinces, Salta stands out as having more effectively implemented reforms, especially in health. In education, all three provinces implemented the legally mandated elements of the reform [increasing the supply of schooling to accommodate the expanded age range for compulsory school attendance, and reorganizing the school cycles and adopting curriculum reform, at least in name]. However, reforms to raise quality were not equally well implemented. 5.4 The Findings and Recommendations. Context: * Both reforms were introduced in a difficult fiscal context for the provinces, but the effects on implementation depended on the provincial response to fiscal crisis. While salary arrears provoked labor unrest, resolving the arrears problem had the potential to create a working partnership between government and labor. * Perhaps more important than the difficult fiscal context as an impediment to reform is the widespread perception on the part of provincial officials of government corruption, which is correlated with ineffective government. This highlights the need for reforms to be considered in the context of broader reforms to modernize the State, increase transparency and accountability, and create the incentives for improved performance of the overall administration. Formulation of Reform: * Broad national debate involving the participation of many stakeholders contributed to the relative success of the education reform and the lack of debate contributed to the lack of understanding and support for the health reform. * Formulation of reform through executive degree has the advantage of not having to wait for the Congress to deliberate and possibly reject a reform proposal, but if the decree is not subsequently followed by legislation the reform runs the risk of never gaining the support and understanding of key stakeholders. 60 Reform Design: * Reforms which the provinces are mandated to put in place are implemented more quickly than those which are voluntary, but those mandates are meaningful only when their implementation is carefully defined and monitored. Because legal mandates focus the energies of the provinces, it's important that they be used for the highest priority elements in the reform. The use of mandates in the education reform may explain the relative slowness in implementing those parts of the reform-- quality- enhancement and teacher training -- which were not mandated. * Complex reforms inevitably run up against constraints in the capacity of the provinces to implement everything at once. A complex reform, like that in education, should provide guidance as to the sequencing of the reform, consistent with provincial capacity to implement. Federal Implementation Strategy: * A significant difference in implementation strategy between the Health and Education Ministries concerned how the minister worked with the Federal Council of Provincial Ministers. The education experience shows how the Council can be a useful vehicle for dialogue and information sharing between provinces and between the provinces and the Federal Government. * Federal financial assistance can be used as an incentive to the provinces to implement reforms and was used as such in both the education and health reforms, but funds for hospital renovation were not closely tied to implementation of hospital autonomy and thus had little effect. * Communication is a key part of any implementation strategy and appears to have been a weak link in both the education and, especially, health reforms. The Education Ministry did use mass media to communicate its reform, but it appears to have not succeeded in bringing parents and teachers on board. The hospital autonomy reform was poorly communicated, with the result that even well-educated stakeholders remained ignorant and unconvinced of the need for reform. * Detailed planning of the implementation is another factor in the Ministry's strategy. Drilling down to determine what has to happen for implementation to take place is a useful tool for identifying constraints and blockages to reform. However, the experience with the health reform shows that good strategic planning is not sufficient to bring about implementation. * The Federal government has the option of trying to bring about change by either working through the provincial ministries or providing direct assistance to the service providers. Direct assistance to schools was perceived as being more effective than assistance via provincial ministries. 61 * The Federal government has a critical role to play in ensuring equity, both inter- regional and inter-personal. As shown in the case of direct financial assistance to schools, the Federal Government can play this role in a way that, also, helps facilitate the implementation of reform. 5.5 Ministry capacity to manage change is another crucial detenninant of success in implementation, and the results of the evaluation of ministry capacity are not very positive. Provincial ministry capacity is quite low across the sectors and across the provinces. Capacity at the level of the service provider may be even weaker. Leadership. * Leadership and commitment at the very top are critical for the implementation of reforms in the provinces. Salta is an example where a governor's leadership resulted in ministry officials being committed to his reform agenda. When a governor offers leadership, he/she may use the powers of political appointment and influence in the bureaucracy to ensure compliance with and support for the agenda. Participation, Commitment and Ownership. - In general, provincial ministries lack the kinds of incentives [regular staff evaluation, meritocratic promotion] and internal communication required for the construction of a sustained commitment to implement reform. - Even reforms which were formulated in a participative fashion may lack ownership when it comes to implementation. The education reform, which enjoyed considerable participation by some stakeholders in its national formulation, is perceived as a top- down reform by teachers and parents. • All reforms face the risk of creating both winners and losers among key stakeholders with the accompanying risk that the losers may effectively block implementation efforts, or at least fail to actively support them, unless those losers are adequately compensated. In the case of education, teachers saw costs and few compensating benefits to the reform. In the case of health, physicians saw risks and were generally uninformed about the potential benefits of hospital reform. Authority * Provincial governments have ample authority to implement reforms even though there may be areas in which respective responsibilities need to be worked out or mechanisms for collaborating across levels instituted. Moreover, they are not good at delegating that authority downwards to program managers, school directors, or hospital directors. * Hospital and, especially, school directors have almost no authority with which to bring about improvements in school quality. This lack of authority [and accompanying financing] constrains problem-solving at the local level. 62 Incentives. * Incentives for either individual staff or ministries or schools or hospitals to perform well and to contribute to the province's reform goals are weak, excepting in cases like Salta where there is alignment between the reform goals and the governor's political goals. * The lack of substantive evaluation of staff in ministries, schools and hospitals means even professionals who wish on intrinsic grounds to improve their performance have no information to act on. * A lack of rewards for high performance combined with a lack of financing for reforms at the level of the service provider seriously constrain attempts to raise quality. In education, these two factors deter faculty and parents from participating in the formulation and implementation of school-level reforms. Accountability. * The lack of information on performance, standards of expected performance, and positive or negative consequences for performance translate into a lack of accountability at the level of the ministry, the school, and the hospital. * A key role of the Federal government is to ensure that information on the performance of schools and hospitals is provided and widely disseminated. The creation of a student evaluation system became the cornerstone for constructing a system of accountability in education. A similar tool is needed for health. Resources The provincial governments have provided the financial resources requires for at least the legally mandated elements of the education reform, but those resources still remain unpredictable, which affects the efficiency of their use. B. LESSONS LEARNED FROM EDUCATION AND HEALTH 5.6 Argentina's experience with education and hospital reform provide lessons to guide Federal and provincial implementation strategies. What lessons should future reformers keep in mind in advancing on the continuing agenda for education and health or for designing and implementing future social sector reforms in other areas. Some of these lessons learned include: * The Federal ministry should work with the council of provincial ministers to build support and consensus prior to announcing the reform and, once the reform is legislated or decreed, again work with the council to build support for its implementation, to discover implementation problems that need resolution, and to facilitate provinces learning from each other's experience. 63 * To ensure acceptance and support for the reform, the Federal ministry as well as the provincial ministries should undertake publicity campaigns to inform beneficiaries of the reform, the benefits to them, their role in the reform (in the case of food stamps, they need to register), and what they should expect to receive from service providers. * To build provincial ownership of reform, reformers should adopt a flexible reform design that allows provinces to adapt policies to their own needs, and build in continuous communication between technical staff at the provincial and Federal levels to ensure provincial staff that their voices are heard. * Enhanced participation of beneficiaries-parents, communities, associations, etc-through out the process can add not only to generating commitment but can also act as a form of social auditing and control on the results of the reforms. They can help to inform policy-makers if the changes are really happening on the ground. * Provide technical and financial assistance to provincial governments and service providers to facilitate implementation of reform. Give the recipient of this assistance some say in what they receive by, perhaps, providing them a menu from which they can choose, or encouraging them to initiate their own requests. * Once the design of the reform is known, the Federal ministry should operationalize the reform by drilling down from the goals to determine where are the constraints that might impede implementation and to determine what assistance the ministry might provide to relax those constraints. The Federal ministry should, also, help the provincial ministries to carry out this same exercise. * Carefully monitor and evaluate the implementation of reform in the provinces (and provincial ministries could do so within provinces) and make the results available to the public. * Governors can facilitate reform by giving it prominence in their agendas, making its implementation a priority, providing the financing needed for implementation, and selecting a minister capable of leading change and charging him with that responsibility. * Ministers can facilitate reform through their appointment of individuals to manage it within the ministry, by making implementation the priority goal of the ministry and assigning staff explicit responsibilities and evaluating performance in executing those responsibilities, and by maintaining an open dialogue with key stakeholders, including those opposed to the reform effort. 5.7 The Federal Role. While the provinces are ultimately responsible for implementing reforms, the Federal government will have an obvious, continuing role in 64 facilitating implementation, assessing results, sharing national and provincial experience and promoting a process of continued learning and innovation. Based on experience to date, there are several roles that the Federal Government might play to facilitate the implementation of reforms to raise the quality of service delivery. In particular, the Federal Government has as a major role the stimulation of social demand for improving quality. Corrales (1999) suggests the following as measures to stimulate demand: (a) launch massive information dissemination campaigns, (b) involve potential beneficiaries in reform design and evaluation, and (c) grant greater financial autonomy to local entities. The Federal Government could work with provincial government ministries to facilitate adoption of these measures. Another important role of the Federal Government is to create a continuous reform based on learning from Argentina's own experience, i.e., to create a learning system. Argentina needs to create mechanisms to identify successful provincial reform experiences that can then inform other provinces. Finally, the generally accepted role of the Federal Government in ensuring inter-regional and inter- personal equity can be carried out in a way that facilitates reform while improving equity. C. CONCLUSION. 5.8 This study covers a small sample-two reforms in three provinces-which argues for caution in generalizing about lessons learned. On the other hand, these lessons confirm implementation experiences reported earlier in other countries. One lesson is to confirm an old one, that leadership counts, especially when that leadership is provided by the governor of a province. A second lesson is that it's difficult to obtain commitment from actors (e.g., teachers, physicians) who have had no participation in reform design. A third lesson is that for Federal reforms to succeed where Federal finance is small and provides little leverage requires a sound communication and information strategy that builds a broad-based demand for reform, or at least diminishes the fear of change. A fourth lesson is to not impose the costs of reform on those who are required to implement it. A fifth lesson is that Federally-initiated reforms are difficult to implement if the Federal ministry itself does not take a proactive stance. A proactive Federal ministry is particularly important when the reform affects sectors -like health and education- which are by and large provincial responsibilities. A sixth lesson is that Federal reforms that are top-down in their implementation face a formidable obstacle in the low effectiveness of most provincial ministries. A final lesson is that the support or opposition of stakeholders can be strongly influenced by the implementation strategy chosen by the province. 5.9 The largest dilemma facing both Federal and provincial governments in implementing reform is how to improve the capacities of provincial ministries. The high degree of political influence and perceived corruption; the lack of relationship between merit and career advancement; the lack of horizontal and bottom-up communication are problems that will require a long time to resolve, and the implementation of reform cannot wait for their solution. Given this weak provincial capacity, there are strong arguments for Federal strategies to provide assistance directly to local service providers, as was done in the Social Education Fund, while simultaneously strengthening provincial 65 capacity, and increasing the expectations and demands by beneficiaries and clients for improved service quality. 66 BIBLIOGRAPHY Abrantes, Alexandre and A. Legaspe. 1999. Contracting Public Health Care Services in Latin America. Paper to be presented at the first Europe and Latin America and the Caribbean forum on health Sector Reform, Paris. May 10-12. Analisis del Grado de Autonomia que Prestan los Hospitales Publicos de la Repuiblica Argentina. 1998. Becerra, Marcelo. 1999. Provincia de Catamarca. Analisis del Gasto Educativo y de la Sustentabilidad de la Reforma - First Draft. November. Bennet, Sara, S. Russell and A. Mills. 1996. Institutional and Economic Perspectives on Government Capacity to Assume New Roles in the Health Sector: Review of Experience. London School of Hygiene and Tropical Medicine. Health Economics and Financing Programme. London, UK. Bennet, Sara, S. Russell and A. Mills. 1996. Institutional and Economic Perspectives on Government Capacity to Assume New Roles in the Health Sector: A Review of Experience, London School of Hygiene and Tropical Medicine. Mimeo. Berrymna, Sue and others. 1997. Guidelines of Assessing Institutional Capability. November. The World Bank. Washington, DC. Bisang, Rigoberto and Centrangolo, Oscar. 1997. Descentralizaci6n de los servicios de salud en la Argentina. Santiago, Chile. CEPAL, Serie de Reformas de Politica PuIblica 47. Braslavsky, Cecilia and G. Tiramonti. 1990. Conducci6n Educativa y Calidad de la Ensefianza. Mifno y Davila editores. Buenos Aires, Argentina. Crouch, Luis and J. DeStefano. 1997. Education Reform Support. Volume Five: Strategy Development and Project Design. February. Office of Sustainable Development. Bureau of Africa. U.S. Agency for International Development. Washington, DC. Direcci6n de Educaci6n Media, Tecnica y Agraria. 1999. Programa de Fortalecimiento Educativo Institucional - ProFEI-. Buenos Aires, Argentina. Direcci6n General de Cultura y Educaci6n. 1999. La Transformaci6n Educativa en la Provincia de Buenos Aires. Provincia de Buenos Aires, Argentina. Direcci6n Nacional de Evaluaci6n. 1994. "Escuelas Efectivas" . Buenos Aires, Argentina. 67 El Presente Argentino Garantiza el Futuro de los Aprendizajes. 1999. Ministerio de Cultura y Educaci6n Presidencia de la Naci6n. Est6banez, Maria Emilia, P. Feliu, and M. Mancebo. 1998. Evaluaci6n Social, PRODYMES. Provincia de Buenos Aires. Experton, William, E. Vega and L. Pritchett. 1999. C6mo atraer y retener docentes calificados en la Argentina: impacto del nivel estructural de la remuneraci6n. April. LCSHD Paper Series. The World Bank. Washington, DC. Filmus, Daniel. 1996. Estado, sociedad y educaci6n en Argentina de fin de siglo: Proceso y desafios. Buenos Aires, Troquel, Fiszbein, Ariel. 1999. Institution, Services Delivery and Social Exclusion: A Case Study of the Education Sector. October. The World Bank Institute. Washington, DC. Fondo Participativo de Inversi6n Social - FOPAR - 1996. Primer Concurso de Proyectos en los Barrios de Salta. Provincia de Salta. Fundaci6n Carlos Auyero. 1998. Plan Social para Erradicar la Indigencia. Buenos Aires, Argentina. Fundaci6n de Investigaciones Econ6micas Latinoamericanas. 1999. La Distribuci6n del Ingreso en la Argentina. Sintesis elaborada para su presentaci6n en la Reuni6n Anual '99 organizada por la Asociaci6n de Bancos de la Argentina (ABA). Junio. Buenos Aires, Argentina. Gobiemo de la Provincia de Buenos Aires. 1999. Programa de Transformaci6n de la Educaci6n de Nivel Medio, 1998. Gobierno de la Provincia de Buenos Aires. Direcci6n General de Cultura y Educaci6n. Direcci6n de Educaci6n Media, Tecnica y Agraria, Unidad Ejecutora Provincial. August. Gonzalez Prieto, Guillermo. 1998. Analisis del grado de autonomia que presentan los hospitales puiblicos de la Repuiblica Argentina. Buenos Aires. Mimeo. Haggarty, Luke and Y. Matsuda. 1999. Assessing Clients' Commitment Sectoral Reformn: A Reform Readiness Analysis. The World Bank. Washington, DC. Harding, April, A. Preker and N. Girishanakar. 1999. The Economics of Private Participation in Health Care: New Insights from Institutional Economics. August. The World Bank. Washington, DC. Harding, April and A. Preker. 1999. Innovations in Health Care Delivery: Organizational Reforms Within the Public Sector. A Conceptual Framework. February. 68 Jaim Etcheverry, Guillermo. 1999. La Tragedia Educativa. Noviembre. Fondo de Cultura Econ6mica. Buenos Aires, Argentina. Kaufmann, A. Kraay and Zoido-Lobaton, Pablo. 1999b. Governance Matters. World Bank Policy Research Department. Working Paper No. 2196. La Reforma del Sector Piublico en Catamarca. 1999. Diagn6sticos, problemas y desafios de la transformaci6n hacia el Siglo 21. Llach, Juan Jose, S. Montoya y F. Roldan.. 1999. Educaci6n para Todos. Noviembre. IERAL. Regional Comahue. Provincia de C6rdoba, Argentina. Llach, Juan Jose y S. Montoya. 1999. En Pos de la Equidad. La pobreza y la distribuci6n del ingreso en el Area Metropolitana de Buenos Aires: digan6stico y alternativas de politicas. Junio. IERAL. Regional Comahue. Buenos Aires, Argentina. Medici6n del Impacto de los Proyectos Innovadores en la Jurisidicci6n Provincial. 1998. Ministerio de Cultura y Educaci6n de la Naci6n. 1997. La Educaci6n Argentina en la Sociedad del Conocimiento. 29a. Reuni6n de la conferencia General de la UNESCO en Paris, 21 de octubre al 12 de noviembre. Repuiblica de Argentina. Ministerio de Cultura y Educaci6n. 1996. Diagn6stico de Relevamiento de la Gesti6n Administrativa de la Provincia de Catamarca. Programa de Reforma de la Gesti6n Administrativa de los Sistemas Educativos Provinciales. Mayo. Argentina. Ministerio de Cultura y Educaci6n. 1996. Diagn6stico de Relevamiento de la Gesti6n Administrativa de la Provincia de Salta. Programa de Reforma de la Gesti6n Administrativa de los Sistemas Educativos Provinciales. Abril. Argentina. Ministerio de Economia de la Provincia de Buenos Aires. 1999. Analisis de los Sectores Sociales del Gobierno de la provincia de Buenos Aires: Salud, Educaci6n y Programas Sociales Focalizados. Descripci6n Diagn6stico. Principales Conclusiones y Recomendaciones. Buenos Aires. Ministerio del Interior y Ministerio de Economia y Obras y Servicios Puiblicos. 1999. Propuesta de Reforma del Regimen de Coparticipaci6n Federal de Impuestos. Buenos Aires, Argentina. Montoya, Silvia y J. Colina. 1999. La Reforma de las Obras Sociales en Argentina. Este trabajo fue realizado como parte del proyecto: Procesos de reforma en educaci6n y salud. Disefio, implementaci6n y grupos de interes" patrocinado por el BID y CIID y coordinado por CIDE e ILADES. Pacto de la Calidad Educativa. Documento del nuevo gobierno provincial a cargo del Dr. Jose Manuel de la Sota. 69 Paqueo, Vicente and K. Lee. 1999. Argentina's Social Sector Development and Poverty. Background paper for "Poor people in a Rich country: A Poverty Report for Argentina". June. The World Bank. Washington, DC. Peirano, Claudia (coordinadora) y otros. 1999. Educaci6n para todos. Situaci6n en Misiones. Noviembre. IERAL. Regional Comahue. Provincia de C6rdoba, Argentina. Poppe de Mufioz, Flavia. 1999. Revisi6n de los Gastos en Salud. Provincia de Catamarca. Presidencia de la Naci6n, Banco Mundial, y Centro Nacional de Organizaciones de la Comunidad. 1998. Programa de Capacitaci6n para las Poblaciones IndIgenas de Argentina. Buenos Aires, Argentina. Programa de Promoci6n de la Reforma Educativa en Arnerica Latina y el Caribe. 1998. Proyecto Sindicalismo Docente y Reforma Educativa en America Latina. Octubre. Boletin No. 1. Buenos Aires, Argentina. Proyecto IERAL - OSDE. 1999. Revista ZONA EDUCATIVA. 1996. Repuiblica de Argentina. Ministerio de Cultura y Educaci6n de la Naci6n. Affo 1 - No. 9. Noviembre. Revista ZONA EDUCATIVA. 1996. Republica de Argentina. Ministerio de Cultura y Educaci6n de la Naci6n. Afio 1 - No. 7. Septiembre. Secretaria de Desarrollo Social de Salta. 1996. Proyecto de Comedores Infantiles a Centros de Cuidados Infantil. Capacitaci6n para 10 Centros de Cuidados Infantil y 10 comedores Infantiles. Programa Alimentario Provincial - PAP-. Salta. Secretaria de Prograrnaci6n y Evaluaci6n Educativa. 1998. La Evaluaci6n una Herramienta para Mejorar la Calidad de la Instituci6n. Ministerio de Cultura y Educaci6n de la Naci6n. Buenos Aires, Argentina. Secretary of State of International Finance. 1999. Province of Salta. Shenone, Osvaldo H. 1987. The Argentinean Experience with the Value-Added Tax. Centro de Estudios Macroecn6icos de Argentina. February. Development Research Department. Economic and Research Bank. The World Bank. Washington, DC. Report No. DRD235. Tappata de, Anahi y R.A. Castro. 1999. Educaci6n para todos . El caso de la Educaci6n Neuquina. Noviembre. IERAL. Regional Comahue. Provincia de C6rdoba, Argentina. The World Bank. 2000. Argentina: Poor People in a Rich Country. Washington, DC. Report No. 19992-AR. 70 The World Bank. 1999. Consultations with the Poor. Prepared for the Global Synthesis Workshop. September 22-23, 1999. Poverty Group. PREM. Washington, DC. The World Bank. 1999. Decentralization of Health Services in Latin America Countries: Issues and Some Lessons. The World Bank. 1997. Second Provincial Reform Adjustment Loans to the Argentine Republic to support four Reforning Provinces: Rio Negro, Salta, San Juan, and Tucuman. July. The World Bank. Washington, DC. Report No. P-7136-AR. The World Bank. 1997. Argentina: Facing the Challenge of Health Insurance Reform. May. The World Bank. Washington, DC. Report No. 16402-AR. The World Bank. 1997. Argentina: Improving Health and Education in San Juan. September. The World Bank. Washington, DC. The World Bank. 1997. Argentina: Improving Health and Education in Tucuman. September. The World Bank. Washington, DC. The World Bank. 1997. Argentina: Improving Health and Education in Salta. September. The World Bank. Washington, DC The World Bank. 1996. ARGENTINA. C6rdoba - Public Sector Assessment: Proposals for Reform. May. The World Bank. Washington, DC. Report No. 15132-AR. Volumes I and II. The World Bank. 1995. Argentina: Argentina's Poor: A Profile. July. The World Bank. Washington, DC. Report No.13318-AR. The World Bank. 1995. Argentine Republic: Decentralization and Improvement of Secondary Education and Polymodal Education Development Project (DISEPED). December. The World Bank. Washington, DC. Report No. 14559-AR. Winkler, Donald. 2001. Technical Document for the Assessment of Institutional Capacity for Social Sector Reform in Argentina. The World Bank. Washington, DC. www.gob. gba.gov.arg/cdi/aspectos%20Sociales/Educacion.html Descripci6n de los programas de educaci6n en la provincia de Buenos Aires. www.gob.gba.gov.arg/cdi/Aspectos%20Sociales/Salud.html Descripci6n de los programas de salud en la provincia de Buenos Aires. www.gob.gba.gov.arg/cdi/Aspectos%20Sociales/ASOCIAL.HTML Descripci6n de los programas de asistencia social en la provincia de Buenos Aires. 71