F'eport No. 24281-AR Argenti na TFogether We Stand, Divided WVe FaIi ILevels and Determinants of Social Capital in Argentina MAay 31, 2002 F'overty Reduction and Economic Management Unit mrgentina, Chile, Paraguay and Uruguay Country Management Department Latin America and the Caribbean Region 9Document of the World Bank GOVERNMENT FISCAL YEAR January 1 - December 31 CURRENCY EQUIVALENTS Currency Unit = Argentine Peso US$1.00 = AR$1.0 (to January 1, 2001) US$1.00 = AR$3.56 (as of June 17, 2002) ACRONYMS AND ABBREVIATIONS PNUD Programa de las Naciones Unidas para el Desarrollo BID Banco Interamericano de Desarrollo CEDES Centro de Estudios de Estadoy Sociedad CSO Civil Society Organizations FLACSO FacultadLatinoamericana de Ciencias Sociales CENOC Centro Nacional de Organizaciones de la Comunidad SCAT Social Capital Assessment Tool WVS World Values Survey NGOs Non-Governmental Organizations FOPAR Fondo Particip ativo de Inversidn Social Vice President: David de Ferranti Country Director: Myrna L. Alexander Sector Manager: Emesto May Country Sector Leader: Paul Levy Task Team Leader: Katherine Bain Table of Contents EXECUTIVE SUMMARY ..............................................................I Introduction ...............................................................i The Reasons for Concern ................................................................i The Main Findings ..................!.............................................£..ii Participation ................................................................ii Collective Action ............................................................... iii Interpersonal Trust ............................................................... iv Social Capital, Trustworthiness, and Institutional Quality ............................................................... iv Moving Forward: Some Recommendations ............................................................... iv I . INTRODUCTION ..............................................................1 The Challenge in Argentina ...............................................................1I The Importance of Social Capital ................................................................2 Studying the Case of Argentina ................................................................2 The Study's Structure ................................................................3 H. THE CONCEPTUAL FRAMEWORK ..............................................................5 The Definitions ................................................................5 Getting a Grip on the Concept ................................................................5 The Benefits of Social Capital in Poverty Reduction and Sustainable Development ..........................................8 Putting Social Capital to Work ................................................................9 Promoting Social Capital: Civil Society Organizations ............................................................... 10 The Role of Government ................................................................11 HI. A BALANCE SHEET ON SOCIAL CAPITAL IN ARGENTINA ............................................................. 13 Taking Stock ............................................................... 13 A Historical Perspective .IS......;1 At What Stage is Social Capital Now? ..................................................... 19 IV. A NEW QUANTITATIVE ANALYSIS ................................................... 21 Description of the Tool ...................................................... 21 Consistency and Improving the Measurement ..................................................... 22 The Survey Results ..................................................... . 22 A. Levels, Types and Determinants of Participation ............................................................ 23 B. Collective Reponses to Shocks ............................................................ 32 C. Levels and determinants of inter-personal trust ............................................................ 35 D. A Baseline for Participation and Trust by Region ............................................................ 41 E. Enabling Environments and Social Capital ............................................................ 44 V. CONCLUSIONS AND RECOMMENDATIONS ................................................... 47 The Main Conclusions and Interpretations ..................................................... 47 A. Social Capital is Low ............................................................ . 47 B. Associations are often Religious Based .............................................................. 48 C. The Family Counts the Most ............................................................ 48 D. Regional Differences are Important ............................................................. 48 E. These Differences Need to be Better Understood ............................................................ 48 F. L inks with Poverty, Gender and Education ............................................................ 49 G. The Poor Participate to Survive while the Rich Participate to Advance ........................................... 49 H. The Poor Don't have a Chance to Lead, Control or Make Decisions ............................................................ 49 1. Trusting Each Other and Trusting the State ................................................. 50 A Broad Agendafor the Future .............................................. 50 A. Creating an Enabling Environrent for Social Capital ................................................. 50 B. Promoting Social Capital ................................................. 52 Moving Forward .............................................. 53 END-NOTE ...........................................55 Discussions on the Report with Social Actors in Argentina ............................................. 55 BIBLIOGRAPHY .......................................... 59 List of Boxes Box 1: Summary of the Voices of the Poor in Argentina ...........................................................................2 Box 2: Social Capital and Other Forms of Capital ........................................................................... 6 Box 3: Social Capital and the World Bank ...........................................................................8 Box 4: Moving from Hostile State to Benevolent State: Increasing the Odds of Social Development .... 12 Box 5: Chronological Synthesis .......................................................................... 16 Box 6: The Urban Movement: Neighborhood Associations in Argentina ................................................ 17 List of Figures Figure 1: The Relationship between Micro and Macro Levels of Social Capital Participation and Confidence in Public Institutions ...........................................................................8 Figure 2: Probability of Membership and Log Per Capita Income ........................................................... 25 Figure 3: Types of Organization in Which Argentines Participate ........................................................... 29 Figure 4: Average Monthly Income of Members of Homogeneous and Heterogeneous Organizations ..30 Figure 5: Average Monthly of Income of Members of Linking and Bonding Organizations .................. 30 Figure 6: Average Montly Income of Members Who Participate in Selection of Leaders ....................... 31 Figure 7: Average Monhly Income of Members Who Participate in Fund Management ......................... 32 Figure 8: Probability of Having Interpersonal Trust and Log Per Capita Income .................................... 37 Figure 9: Probability of Having Interpersonal Trust and Wealth ............................................................. 37 Figure 10: Levels of Interpersonal Trust by Participants and Non-Participants ....................................... 39 Figure 11: Levels of Interpersonal Trust for Participants in Homogeneous and Heterogeneous Organizations .......................................................................... 40 Figure 12: Levels of Trust of Members that Participate in the Selection of Leaders ................................ 40 Figure 13: Trust and Participation by Region .......................................................................... 43 Figure 14: Levels of Macro Trust for Participants and Non-Participants ................................................. 44 Figure 15: Levels of Interpersonal Trust and Trust in Public Institutions ................................................ 45 List of Tables Table 1: Rates of Poverty, Extreme Poverty .........................................................................1 Table 2: Confidence that Certain Institutions can Resolve Social Problems ........................................... 13 Table 3: World Values Survey: Levels of Participation and Trust .......................................................... 14 Table 4: Non-Governmental Organizations by Province ........................................................................ 15 Table 5: Characteristics of Individuals Who Participate ........................................................................ 23 Table 6: Determinants of Membership, Various Specifications .............................................................. 26 Table 7: Participation Rates by Regions ........................................................................ 27 Table 8: Reasons Offered by Respondents for Participating in an Organization ...................................... 27 Table 9: Characteristics of Individuals Who are Leaders ........................................................................ 28 Table 10: Participation Rates by Internal Characteristics of Organization .............................................. 31 Table 11: Assistance in the Case of Job Loss ........................................................................ 33 Table 12: Assistance in the Case of Job Loss by Employment and Unemployment Groups ................... 33 Table 13: Sources of Expected Relief in Systemic Shocks ..................................................................... 34 Table 14: Frequency with which Members of the Community have Presented a Joint Petition to Government Officials or Political Leaders ........................................................................ 34 Table 15: Level of Contact with Influential People ........................................................................ 35 Table 16: Reasons for Not Participating in Petitions ........................................................................ 35 Table 17: Trust with Participation Rates by Regions ........................................................................ 36 Table 18: Characteristics of Individuals Who Have Interpersonal Trust ................................................. 36 Table 19: Determinants of Interpersonal Trust, Various Specifications ................................................... 38 Table 20: Predisposition to Contribute to a Project from Which One Does Not Benefit Directly .......... 41 Table 21: Trust with Participation Rates by Percentage of Population by Region .................................. 41 This report was prepared by Katherine Bain (Task Manager, LCRVP), Sandra Cesilini (LCC7C), Daniel Lederman (LCRCE) and Norbert Schady (LCSPP). The valuable comments of Myrna Alexander, Shelton Davis, Paul Levy, Roby Senderowitsch, Christiann Grootaert and Michael Walton are greatly acknowledged. Irene Novacovsky (Buenos Aires, Argentina), Gabriela Carrasco and Mariela Alvarez, from the World Bank, provided outstanding support. EXECUTIVE SUMMARY LEVELS AND DETERMINANTS OF SOCIAL CAPITAL IN ARGENTINA lntroduction I . Participation, trust and solidarity are arguably goals in themselves. But these elements of 'social capital" are also means to further valuable ends. An increasing number of empirical studies suggest that social capital can have significant economic and welfare-enhancing benefits, Es well as providing a crucial unifying force that brings citizens and governments closer together. 1These empirical studies provide new and convincing evidence of the-role that social capital can play in promoting steady and equitable growth, improving productivity, providing cohesion in times of strife, serving as a safety net for the poor and vulnerable, protecting communities, promoting acceptable norms, and reducing poverty (Knack and Keefer 1997, Collier 1998, Temple 1998, Easterly 2000, Narayan and Pritchett 1998, Grootaert 2000, Bowles 1999, Seligman 1997, Abraham 1985, Moser 1996). The benefits of social capital lie in the role that both participation and trust play in social interactions. Trhe Reasons for Concern 2. Argentina is a country currently facing enormous challenges. Recent analytical work done by the World Bank suggests that both inequality and unemployment are getting worse, and ihat earlier gains made in the fight against poverty are being lost (World Bank 2000). In addition lo the social problems that Argentina faces, the country is engulfed by a sense of collective ])essimism and despair, in part shaped by a more than 30 month recession. Analytical work and niumerous opinion polls undertaken in Argentina over recent months point to dissatisfaction with public institutions, increasing isolation of citizens, and a sense that further decline is inevitable. [f there was ever a time that Argentina needed to draw on its existing social capital as an Immediate coping strategy and an investment for the future, it is now. Providing cohesion among citizens and between the government and its citizens, as well as harnessing private action to complement public action, is clearly a high priority if Argentina is to rise to the challenges it Faces currently. This report sets out to assess whether, in this context, social capital can play a role. 3. This report does not provide immediate hope. It finds that aggregate levels of social capital in Argentina are fairly low: only 20 percent of the population participate in any informal group or organization and 23 percent trust their fellow citizens. Furthermnore, the social capital of the poor in Argentina is of a type that enables them to protect themselves in times of hardship but does not help them to get ahead in the long-term by linking them to those with power or new resources. While there were clearly signs of civic participation in Argentina prior to the formation of the nation state, civic associations have proved vulnerable to deep changes in the i local social, economic and political landscape. In addition, at different times in the history of Argentina, civic organizations have either been crowded out or co-opted by the state. Thus, at present the low levels of participation would appear to stern, in part, from this heritage of authoritarian and clientelistic relations with the state. As a result, Argentines appear to be hesitant or skeptical about adjusting to a new formn of social contract, characterized by a facilitating state and active citizenship. In addition, the credibility crisis which surrounds public institutions and the broad lack of trust in all that is public presents a significant barrier, and the low levels of participation and trust at the micro level are almost certainly not enough to overcome this. The Main Findings 4. The findings of the report on the levels and determinants of participation and interpersonal trust provide a snapshot of levels of social capital in Argentina The report does not make an exhaustive study of the phenomena of social capital, nor does it provide clear answers. Rather it attempts to provide some evidence which, in turn, could serve to promote a dialogue among national actors on the implications. Participation 5. Less than 20 percent of the population participates in any form of organization or informal network. This level of participation is low, even in comparison with other developing countries. There are also significant regional differences in the levels of participation: both the South and Cuyo have significantly higher rates of participation than other regions. The region including Buenos Aires - the Metropolitan region - has the lowest level of participation. 6. Those Argentines that participate do so in a variety of organizations and groups but religious groups are the single most important type of group or organization in which Argentines participate (47%). There is also a wealth of self-help groups (28%) and social groups. While participation is low in Argentina, those that do participate claim to do so in order to feel more connected with others, a part of something, and useful in contributing to a common good. 1 7. These reasons for participation are to be expected in a country where social disintegration and isolation are perceived to be high. Women and the unemployed are the most likely to participate in order to contribute and feel part of a broader group. This may be due to their degree of marginalization which makes them want to reach out more in an effort to be included. 8. The report identifies several determinants of participation in Argentina. While the findings on income, employment and education must be considered with caution, the findings show that those who are better off, higher educated and unemployed are more likely to 2 f parficipate. Being female, a household head, slightly older, and married are all significant predictors of participation. While the poor are less likely to participate than the significantly 'Some comparative data is provided in Table 2. 2Such findings are potentially affected by endogeneity problems which cannot be controlled for without panel data. the very poor participate as part of a coping strategy while the less poor find the opportunity cost too high or the experience too unrewarding. Although men participate less than women, leaders of organizations are still more likely to be men. 9. Another significant finding is that the internal characteristics of the organizations are important in terms of the levels and type of social capital that they promote. Characteristics of the organizations of the -poor are significantly different from those organizations that the better off tend to join. Poor people are more likely to participate in organizations with those that share their backgrounds, skills and opportunities. Their organizations are more likely to be run in a hierarchical manner and their lack of involvement in the organization's affairs is likely to prevent them from gaining new skills and experience. 10. The poor are also more likely to participate in "bonding" organizations. These organizations play an important role in helping the poor to manage risks and shocks. However, they do not provide the opportunity for the poor to leverage external resources and effect more long-term benefits. 11. Last but not least, the organizations of the poor involve their members less in the running of the organizations. They are more hierarchical and thus participants gain less from their participation than those who belong to organizations with horizontal and inclusive decision- making. The better off, in contrast, are more likely to participate in more heterogeneous groups where they meet people with different information and resources. Their "bridging" organizations are used more to reach out across boundaries. Members of these organizations are also more likely to have some control over the group's decision-making and activities. Collective Action 12. The report also measures the levels of spontaneous or less structured collective action in response to shocks in addition or as a strategy for interacting with public officials. During any form of crisis, Argentines turn to their closest circles of family and friends. The unemployed, the poor and informal sector workers are more likely to feel that they have no-one to turn to and that they face hard times alone. As shocks become more widespread and individual job losses turn into widespread unemployment, Argentines -not surprisingly- feel less supported. External agencies - both public and private - figure low as sources of support during difficult times.. 13. During "normal" times, Argentines seldom engage in collective action. They do not use it as a strategy to engage with public officials or assert their influence on public decisions. This is not, they say, because they do not think it will be effective. Rather they are busy and do not have time or lack information on initiatives; the cost -of dedicating time to a community initiative is considered carefully. However, lack of infolmation among community members and lack of awareness about the benefits of collective action also figure prominently and suggest that getting information into the public domain represents an area for improvement among local organizations. Interpersonal Trust 14. As with participation, levels of trust in Argentina are low compared with other developing and developed countries. Regional differences in levels of trust are significant, as are the regional differences in levels of participation. Both income and wealth seem to be determinants of trust along with education. Household heads are more likely to trust but those who are married are less trusting. Participation is also a significant predictor of interpersonal trust. In particular, horizontal organizations which involve members in the selection of leaders and in management of money promote higher levels of trust than hierarchical organizations. This confirms the hypothesis that the experience of participation is more important than the act of belonging to an organization. 15. Finally, organizations which bring together members with different background tend to create more interpersonal trust than organizations with homogeneous membership. This suggests that organizations which bridge across differences among their membership and provide a space for differences to co-exist promote more trust among their individual members. However, the organizations of the poor are more likely to be homogeneous as well as hierarchical, making it less likely that the experience of participation will result in increased levels of social capital. Social Capital, Trustworthiness, and Institutional Quality 16. Finally, the hypothesis that social capital is inter-dependent on its environment was tested. We find that there does seem to be a significant, synergistic relationship between the two. Argentines who have high levels of confidence both in other citizens in general, and in the effectiveness of their public institutions and their fellow citizens, are more likely to have higher levels of participation and interpersonal trust. However, confidence in public institutions is generally low in Argentina, and particularly low among women, the unemployed, and the poor. This lack of confidence in public institutions and general distrust is, therefore, an obstacle to the development of social capital in Argentina. Moving Forward: Some Recommendations 17. This report is the first step in assessing the levels of social capital and in identifying its determinants in Argentina. To further inform public policy, it recommends that additional qualitative and quantitative analysis be carried out to better understand more precisely the regional differences, the impact of institutional interventions on social capital, and the feasibility of removing certain obstacles. The findings point towards the following five general recommendations: > Creating a Climate that Enables the Development of Social Capital. To create an environment conducive to social capital, mainstreaming public accountability mechanisms in public institutions is recommended. Such mechanisms would promote: information dissemination to the public; civic engagement in the planning of sectoral actions; and user feedback in monitoring the impact and use of public funds. This approach to institutional reform would create a more enabling environment for social capital and provide a way for the state to harness important private and civic initiative. iv Providing Space for Private-Public Interactions. Effective decentralization of key social services in a way that builds the local capacity of municipalities and citizens to interact is recommended to provide a space for a new form of social contract, based on collective but clear responsibilities between citizens and their representatives. Moving away from a culture that expects the state to be the sole provider requires a clarification of public and private roles and responsibilities. Such clarity is best provided at the local level, where citizens and municipalities can interact together around common and achievable goals. > Investing in Citizenship through Education. To ensure that Argentina becomes a country of responsible citizenship and unity among diversity, investments in the education sector are critical. On the one hand, education attainment has a significant effect on participation and, on the other, schools and cooperative school groups are seen as important and trusted organizations in Argentina. Using education to leverage public-private collaboration, and the curricula as a means for promoting awareness of citizenship are two concrete measures. > Creating a Culture of Information Dissemination and Transparency. Creating practices whereby information on the use and impact of public resources is systematically and clearly made available to the public is key to promoting social capital. This implies investing in new practices and communication channels within the public sector and creating capacity for high-quality, independent investigation among private actors. Civic organizations also need to ensure that information on opportunities for participation and, ultimately, the benefits of participatory initiatives are publicized, as Argentines cite lack of information as a reason for not participating. Providing information in an accessible form to the poor and excluded is a general and helpful role that civic organizations might play. > Investing in the Organizational Capacity of the Poor. This report finds that, in a country with little participation, the poor are less likely to participate than the rich and the type of organizations in which they participate - while important as a mechanism for social protection - is unlikely to provide them with significant opportunities for long-term poverty reduction. Both public and private programs must endeavor to: (i) involve the poor in identifying priority actions for any development interventions; (ii) build the capacity of the poor to participate in decision-making, strategic planning, and financial management as an integral part of development projects; and (iii) support the organizations of the poor in extending their connections through learning networking skills, investing in technology that will put them in touch with others, and providing spaces for direct interaction between the poor and those in positions of power. Helping the poor to organize across traditional groups, to leverage new opportunities, and to find collective solutions will more likely to lead to long-term gains in poverty reduction. v IA I. INTRODUCTION The Challenge in Argentina 18. "Poor people in a Rich Country," the title of the recent World Bank poverty assessment, captures the contradictions and challenges Argentina is facing today. Despite its wealth, Argentine is also home to relatively high levels of poverty and inequality. Since 1994 poverty levels have begun to rise: in 1998, 29 percent of the population were living in poverty (World Bank 2000). There is also evidence that inequality and unemployment are on the rise. Table 1 presents the poverty and inequality rates as measured by the World Bank poverty report (World Bank 2000). Table 1: Rates of Poverty, Extreme Poverty (urban areas only) Poverty Extreme Lower Upper 20% / Year Headcount Poverty 20% Lower 20% 1990 41.4 11.3 4.55 11.19 1991 30.4 5.9 4.50 11.42 1992 24.1 4.4 4.85 10.39 1993 21.8 4.3 4.50 11.31 1994 21.6 3.7 4.55 11.23 1995 27.2 6.1 4.30 12.14 1996 30.0 7.3 4.05 13.09 1997 29.4 6.8 4.00 13.30 1998 29.4 7.1 3.80 14.28 Souse: Calculated from EPH surveys, average of May and October. In this Table the poverty rate includes the extreme poor. 19. The economic reforms in the 1990s gave some cause for hope but since then growth has been anything but stable. The benefits of earlier gains have not translated into welfare improvements for the poor, and worrying levels of poverty and inequality are, once again, reason for concern (World Bank 2000). 20. There are signs of social stress and exclusion. Violence and delinquency are on the rise, and several national opinion polls as well as recent qualitative work have captured a collective sense of pessimism, apparent particularly among the poor. Several polls and studies show that the poor do not feel that they are better off now than they were before the reforms, and that their pessimism is shared more broadly by other Argentines (Zuleta and Cesilini, 1999; Patino and Meyer, U.S Embassy Opinion Survey). 21. These studies suggest that there is a cloud of collective pessimism over Argentina, characterized by a fear of what the future might bring, by a feeling that one faces problems alone and is powerless before such challenges, and by a lack of confidence in the capacity of public institutions to help (Voices of the Poor, World Bank 1999, see Box 1). This situation does not provide a context conducive to citizens standing together and uniting with their government to address the growing social divisions in Argentina. Rather, it suggests that Argentina is a country that is divided. Within this climate, the question arises as to whether social capital can play a role in promoting unity and integration, and in enhancing household welfare in Argentina. BOX 1:SUMMARY OF THE VOICES OF THE POOR IN ARGENTINA In 1999 a series of exercises was conducted with poor people in Argentina, as part of the global World Bank initiative, Voices of the Poor. The following excerpts provide a summary of the issues raised by the poor in Argentina. * Growing inequality, on the one hand, and an increased social pressure to keep up with a society that is becoming more consumer- orientated, on the other hand, is also given as a reason for feeling worse off: " .. thepoint Is, the moreyou have, the moreyou need; my kids comefrom school and tell me that they want certain brand-name shoes, clothing.. it's a conswner society... it's a great contradiction" * Low social capital and solidarity leads to perceived loss of collective values and a need to act individualisticallj. This is also an increasing sense of isolation and despair "There are ... people that take advantage... you also can see it in the behavior of potitkians, it's not there... common peoples' worries don't interest them. * In this context, and with the increasing diversity in the population due to immigrant influxes, lack of unity, social cohesiveness and harmony is cited as another reason for further concern. * The consequences of this worsening situation are expressed as a general uncertainty and fear about what the future holds, and an increasing sense of despair. In this situation, maintaining spiritual values are seen to be the only hope in a daily situation characterized by a fierce pressure to "keep up" and-when that inevitably fails-increasing family and social violence, a worsening of health, and a growth in addictions and psychologically-related disorders. * Government at all levels is seen to be corrupt. unresponsive and inefficient. While a strong.role for government is seen as being central to development, local, provincial and central governments are seen as being unconnected and, after having won their vote, unconcerned with the poor. "Arournd here, when people need something, they go to the municipaLity, knock on the door but no one sees them because there isn't anyone there or they havejust gone ow. .. they'U come back tomorrow, maybe the day after..." * "The governor is there because we put him there, we were the ones who voted him in, we're the ones who suffered the heat, the cold, we went thmugh so much to vote so they can get there and today they don't remember us...they take their position and forget about us (the peasants)." "the poUlticiam cheat us, Ut's a corruptness, it's a deception; before they're taught to tal&, they're taught how to make easy monev... " The Importance of Social Capital 22. The contributions of social capital have been documented in recent years. In studies throughout the world, social capital has been shown to play two main roles. First, it has clear economic and social benefits, and is shown to promote sustainable growth (Collier, 1998; Temple, 1998; Easterly, 2000), as well as welfare, productivity and social safety nets (Narayan and Pritchett, 1999; Grootaert, 2000; Bowles, 1999; Abraham, 1985; Moser, 1996). Second, it has less tangible but equally important benefits of cohesion, unity and a sense of belonging and common purpose among citizens - even in hardship (Seligman, 1997; Portes, 1999). Studying the Case of Argentina 23. Both the welfare enhancing and unifying role of social capital offer a potential way forward. Can social capital provide the necessary "glue" to bring Argentines to unite together around a common cause? Or, will the absence of social capital result in a nation that, unable to overcome its divisions and challenges, becomes more divided and less able to find a unified approach to sustainable development? 2 24. This report takes a first step in answering these questions by examining the determinants of social capital. It uses a quantitative survey of 2,235 households to establish a baseline for social capital in Argentina at the regional and national level and provides a framework for future tracking. 25. In doing so, it attempts to answer four general questions: Who has social capital in Argentina? Does the type of social capital of the poor differ from the social capital of the rich? What determines levels of social capital in Argentina? And, what, if anything, can be done to promote social capital in Argentina? The report does not intend to establish causality between poverty or inequality and social capital, nor does it provide evidence of how to promote social capital. While both these questions are of interest in Argentina, they fall beyond the scope of this study and might be better approached with methodologies such as panel data and qualitative research. Finally, this report does not attempt to add to the increasing number of studies on the growing number of civil society organizations, otherwise known as the Third Sector, in Argentina (PNUD, BID 2000 and Johns Hopkins CEDES, 2000) The Study's Structure 26. The report is structured as follows. Section II provides an overview of the conceptual debates surrounding "social capital" and offers a working definition for understanding social capital. It asks why social capital has attracted such attention in development circles. In response, it presents an overview of the evidence on the contributions of social capital to development. Section m takes stock of available knowledge on social capital in Argentina and, based on this, presents some hypotheses for exploration in this report. The fourth section presents the survey tool used for the study and presents the survey's findings on levels of participation, interpersonal trust and general trust in Argentina, and on some respective determinants. In doing so, it provides a baseline study at the national and regional level for future tracking. In the fifth section, the study's conclusions and some broad recommendations are presented. Clearly, both the findings and the directions to take in the future need to be debated and fleshed out. One of the purposes of the study is, in fact, to stimulate discussion of social capital in Argentina. 3 4 II. THE CONCEPTUAL FRAMEWORK The Dermiitions 27. The working definition of social capital adopted for this report is: the networks, associations, norms and values which enable people to act collectively in order to produce a positive externality for them or their community. Thus, this report is concerned with both actions of participation and collective action as well as the values of trust and solidarity. This definition acknowledges many of the conceptual debates surrounding "social capital" and recognizes the potential benefits of social capital at both the individual and the community level. 28. Social capital has three dimensions in performing bonding, bridging and linking roles (World Bank, 2000): * By bringing people with similar characteristics together to solve a common problem within the group, bonding social capital can serve as a collective coping and risk management strategy. * Bridging social capital brings people from different backgrounds together in network- type organizations or broad social movements, such as those that have demanded change most recently in Yugoslavia, the Philippines, and Peru. * Linking social capital provides a connection between individuals with little access to resources and power beyond the community to those organizations that make decisions and control resources. 29. Finally, while it is recognized that the consequences of social capital can be negative, this report is only concerned with its positive use3. The conceptual debate, which broadly covers the qualities, components, benefits, drawbacks, scope, dimensions, and levels of social capital, is elaborated below. Getting a Grip on the Concept 30. As a relatively new concept, there is a debate about what social capital "is". When the term "social capital" is applied to large scale communities such as countries or regions, it seems to state the obvious: countries that have long democratic traditions behave in democratic ways and those that have short or no democratic tradition behave accordingly. At the same time, many have criticized social capital for being a "catch all" phrase with little analytical clarity. The circularity and all-encompassing nature of the concept has brought criticism (Solow 1995). Yet the desire to better understand the role that values, interpersonal relations and organizations and informal networks play in development have encouraged others to continue searching for more conceptual clarity. 3 Social capital can have negative ends as is discussed in Portes and Landolt (1996). Examples include groups such as the Klu Klux Klan, ethnic cleansing groups and the Mafia. Social capital can also be used as a force to constrain the freedoms of certain groups such as those of women by the Muslim fundamentalists in the Middle East. 5 The conceptual debate underlines the qualities that social capital shares with other forms of capital. It depletes, it is renewable, it requires investment and it yields benefits. Like human and productive capital, social capital is thought to be an exponent - that is, it is not sufficient for achieving sustainable development, but performs as a shift factor that enhances or complements other types of capital (Grootaert, 1997). The similarities between social capital and other forms of capital are described by Coleman (1990) in Box 2. Box 2: Social Capital and Other Forms of Capital "Just as physical capital is created by making changes in materials so as to form tools that facilitate production, human capital is created by changing persons so as to give them skills and capabilities....Social capital, in turn, is created when the'relations among persons change in ways that facilitate [collective] action.. .The function identified by the concept "social capital" is the value of those aspects of social structure to actors, as resources that can be used.. .to realize their interests." (Coleman, 1990) 31. In addition to drawing parallels between social capital and other forms of capital, the discussion has focused on five other broad themes. First and foremost is the discussion of the components of social capital. What does social capital comprise? Originally the term was defined as membership in organizations. This led to a number of empirical studies where social capital was measured solely as civic engagement, understood as membership density. More recently, attention has been focused on another, less tangible side of social capital: values, attitudes and norms. Some have argued that a distinction can be made between the "structural" and "cognitive" aspects of social capital. The structural side refers to organizational structures, participation in groups and collective action, while the cognitive side refers to the values of trust and solidarity (Uphoff ,2000). 32. A second theme in the literature concerns the potential benefits of social capital. Below, this study provides a synthesis of empirical work at the country level, as well as cross-country comparative studies. These studies suggest that social capital can make several significant contributions to equitable and sustainable development. More recently, there has also been some recognition of the potential drawbacks of social capital, including the material costs of participation in general, the use of collective action for perverse ends (Rubio, 1997), and the restrictive nature of certain groups (Portes and Landolt, 1997). 33. A third theme addresses the issue of scope: whether social capital is an individual or a community asset. While it was clear that more than one person is needed to build social capital, it was not clear if the benefits of collective action were limited to group members, or if positive "externalities" were generated that benefited non-members as well. There now seems to be some agreement that social capital can benefit many levels of society from individuals to communities (and perhaps even nations as a whole) and that benefits can be enjoyed by communities and individuals simultaneously. 34. The dimensions of social capital is the fourth theme, with important implications for poverty reduction and development strategies. The World Bank's World Development Report 2000/1 defines three dimensions of social capital. Social capital that brings individuals or 6 communities with similar characteristics and opportunities together around common goals has become known as "bonding" social capital (Gittell and Vidal, 1998; World Bank 2000). Examples of this type of social capital include rural women's credit cooperatives and communal kitchens. These examples of bonding social capital serve as strategies for small-scale self-help schemes and risk management. This type of self-help is important - especially to the poor and vulnerable - but does not serve as a force for significant change, since participants share similar problems as well as limited resources, knowledge and contacts. Bonding social capital can be characterized as efforts of like getting together with like to "get by", rather than to "get ahead" (Briggs, 1998). The second dimension is "bridging" social capital, which connects individuals with broadly comparable economic status and political power, but who reside in different geographical areas, have different types of occupations, and thus different social networks. A third dimension is 'linking" social capital, which consists of the vertical ties between poor people and people in positions of influence in formal organizations. This additional dimension is of importance to poor communities, who are usually excluded from the places where major decisions relating to their welfare are made. For the poor and other excluded groups, who face sirnilar problems of lirnited access to public information, powerful contacts and other resources, linking social capital is thought to have a more significant, long-term poverty reduction role than bonding or bridging social capital. Linking social capital helps people to "get on" and not just "get by". 35. The fifth theme is the relationship between micro levels of participation and inter- personal trust, and broader issues such as governance, society-wide confidence and trust in public institutions. Early work on social capital focused primarily on civic engagement and inter- personal trust at the community level. However, more recent work has recognized that micro levels of social capital do not develop in political or institutional vacuums. Recent studies point to the importance of macro variables such as levels of public confidence in public institutions, rule of law, effectiveness of governance and service delivery, and history of social relations. 36. These macro variables are considered key in providing environments which are either conducive to the development of social capital or represent an obstacle to its formation at the local level. Isham and Kaufman (1999) find that weak and indifferent governments have a profoundly different effect on social capital than govemments which uphold civil liberties, rules of law and transparency. Tendler (1994, 1997), Skocpol (1996), Fox (1996), Alesina and la Ferrara (2000), and Brehm and Rahn (1997) present evidence suggesting that there is indeed a complementary relationship between high levels of civic engagement and interpersonal trust at the micro level and a strong and facilitating government at the macro level. Evans (1996) argues that not only is there synergy between public and social institutions, but macro conditions-- a favorable climate-- are crucial to the scaling up of micro level social capital so that it is "politically and economically efficacious." 7 37. Although causality is Figure 1. The Relationship between Micro and Macro not established by any of these Levels of Social Capital Participation and Confidence studies, the relationship in Public Institutions. between micro levels of social capital within communities and macro concerns of trust Trust and confidence more broadly across society is generally believed to be strong and two- way, with each level nurturing the other. Brehm and Rahn (1997) present the links |Participation < Confidence in between macro and micro Public Institutions levels in a diagram in Figure 1. The Benefits of Social Capital in Poverty Reduction and Sustainable Development 38. The primary interest in assessing social capital is to be able to use better this dimension of the development paradigm to enhance poverty reduction and sustainable development. Certain forms of social capital can be used as a part of! a coping strategy for helping people to work together to address common problems, while other forms of social capital can be deployed as a development tool for helping people connect with those beyond their traditional groups. While social capital is primarily a micro level phenomena, recent evidence suggests that broader macro concerns that shape the institutional context for social capital are also key. Such concerns include providing the rule of law, civil liberties, effective and transparent governance, and public promotion of private, civic participation. Box 3 presents a brief summary of the World Bank's growing interest in social capital. Box 3. Social Capital and the World Bank As in other development organizations, the contribution and operational implications of promoting social capital have gained much attention in the World Bank in recent years. This new attention is due to three factors. Fist, the World Bank's broadening concept of wel-being has shifted attention beyond economic causes to a web of inter-related causes (World Bank, WDR 2000). Second, there is a realization that institutional contexts, communities and relationships are important for development (World Bank, 1998, 1999). Third, the World Bank has adopted a broad definition of "institutions" which are defined as the rules that shape the behavior of organizations - private and public, formal and informal, and individuals in society (World Bank, 1999). Thus, new attention is being given to private and informal forms of collective action, and in particular to the role of civic organizations. The general rise in interest in social capital outside the World Bank, then, coincides with these three developments within the institution. Subsequently there has been a series of efforts and empirical studies, including the Social Capital Initiative, which funded a series of case studies and empirical studies from around the World. As a result, the World Bank has compiled evidence suggesting that: (i) social capital is a potentially significant asset for the poor, (ii) social capital can have a strong and positive impact on poverty reduction projects as well as on the broader development context and, (iii) it is lkely that it can be promoted and/or prevented by the World Bank's shareholders and clients. It therefore continues to be an area of analytical and operational interest. 8 39. Perhaps the most recent and exciting evidence on the benefits that social capital can make to poverty reduction and sustainable, equitable growth is found in the growing number of quantitative cross-national studies. Knack and Keefer pioneered this approach and found that both participation and trust have significant, positive impacts on aggregate economic activity (1995, 1997). Collier and Gunning (1999), in their analysis of slow growth in Africa, also find a correlation between slow growth and low levels of social capital (see also Collier 1998, Temple 1998). Easterly reports that countries that are able to draw groups together and to generate consensus are more likely to produce stable and positive rates of growth (2000). While methodological questions can be raised, the overwhelmning message is that civic engagement, trust, effective institutions and enforcement of rights can all provide major contributions to prosperity. 40. Recent country studies have also found both a significant impact of social capital on poverty and a disproportionate value of social capital as an asset of the poor. Narayan and Pritchett (1999) demonstrate that possession of social capital by households in Tanzania has surprisingly strong effects on household welfare. Indeed, the magnitude of the estimated effect is far greater than that of education and physical assets. A study of local level institutions (financed by the World Bank) finds that social capital is a particularly valuable asset for the poor in Burkina Faso, Bolivia and Indonesia (World Bank 1999). Using econometric models, Grootaert finds that the returns on social capital are relatively higher for the poor than the rich. In addition, he concludes that: (i) while there are spillover effects at the community level, individual household actions are required to maximize the effects of participation, (ii) the experience of participating in key tasks such as decision making within organizations is important, and (iii) the composition of the group's membership is also significant to the number of benefits bestowed on members. Putting Social Capital to Work 41. There are at least four types of contributions that social capital can make to development. First, participation can play an important role in self-help and risk management.4 By bringing individuals together around a common and shared goal, participation can serve as a way to achieve ends beyond their individual means. Initiatives that encourage people to participate, pool collective risk in community credit schemes, and/or produce a comrnon good (such as food for neighborhood children, or the rotation of child care responsibilities between community members) are all examples of participation as a means for accessing new resources, meeting basic needs, and providing a safety net. Such initiatives have been well documented by Moser (1996), Narayan (1998), Bowles (1999), Abraham (1985), and others. 42. Second, local participation can have a positive impact on development interventions. The World Bank and other development agencies have measured and documented the clear benefits of promoting participation on a project's design and implementation (World Bank, 1998). Participation tends to make projects more suitable to local needs, ensures that investments are sustained and that public resources are managed effectively and transparently. This is attributed to the fact that collective action can, on the one hand, give "voice' to otherwise 4 Participation is used to refer to regular interaction in a structure to achieve a common goal. Collective action refers to a more spontaneous and less structured interaction to achieve a common goal. 9 excluded stakeholders and, on the other, provides important checks and balances on government action (World Bank 2000). 43. Third, participation and collective action can help communities to maintain their own standards and curb narrow self-interest by prescribing norms and values and exerting some social control. Sampson et al (1999) documents how social capital achieves "collective efficacy" in some Chicago neighborhoods where neighbors speak sternly to youngsters skipping school, or indulging in graffiti and, in doing so, enforce certain behavior. Controlling for a wide range of community and individual characteristics including race and past crime rates, violence is markedly low in these communities and well-being is perceived to be higher. In a similar vein, Coleman (1990) illustrates how external and internal sanctions associated with norms of trust, constrain unproductive opportunism. 44. Fourth, participation helps build trust and a sense of cohesion in communities (Seligman 1997). Thus, participation tends to lead to more trust and a sense of belonging and unity among citizens. This sense of community can be a powerful force for solidarity - especially when a community faces economic hardship, conflict, or other hardships (Chambers 1997). 45. The contributions of trust to development have also been documented (Fukayama, 1995; Knack and Keefer, 1997). Trust can facilitate exchange, lower transaction costs, reduce the costs of information, permit trade in the absence of a contract, and encourage responsive citizenship and collective management of resources (Fukayama, 1995). Interpersonal trust can make societies less dependent on formal institutions; similarly, when formal institutions are weak, trust can act as a temporary substitute for enforcement of agreements and for information sharing. Trust can promote information sharing and mutual knowledge about future states of the world and the response of other agents. Coleman (1990) illustrates this with the example of diamond merchants who trade millions of dollars worth of diamonds with a mere handshake. Trust also promotes higher productivity (Knack and Keefer, 1997), more returns to education (Barro and Lee, 1993; Coleman, 1988), and even raises the probability of citizens voting (Knack, 1992). Finally, governments in societies with higher aggregate levels of trust are likely to be seen as more trustworthy and their words as more credible. Thus assurances by ministers of finance on exchange rates and central bank officials on interest rates are more likely to provide credible assurance and horizons for investor decisions (Knack and Keefer 1995). Promoting Social Capital: Civil Society Organizations 46. Since the potential of social capital has become increasingly well accepted, development organizations have been keen to understand how it can be promoted, what (or who) prevents it, and how barriers or obstacles to building social capital can be removed. In their effort to better promote social capital, development organizations initially turned their attention to the role of civic organizations. These organizations were thought to be the primary promoters of social capital, as they both facilitated participation while strengthening democratic traditions and values. 47. While civil society organizations (CSO) are clearly not synonymous with social capital, they are considered to have two effects in promoting social capital. First, the democracy 10 strengthening potential of civic organization has most recently been illustrated in Yugoslavia. This potential of civic organizations is attributed to their ability to promote social change by creating broad coalitions for change, and promoting new civic values and the democratic rule of law during and after transitions (O'Donnell, 1992 and 1994; Smillie, 1995; Korten, 1990). Civic organizations are also thought to have a second potential comparative advantage in promoting collective action-especially among excluded populations. This comparative advantage is thought to lie in the fact that they are often small, flexible and decentralized organizations that work closely with poor members of communities and practice participatory methods, which in turn help foster collective action (Clark, 1991). Thus civic organizations are broadly viewed as being both engines for collective action and fuel for the creation or enforcement of democratic environments. 48. Not all civic organizations can or do play these roles. Even those organizations which are dedicated to promoting civic action and poverty reduction face difficulties. Some local civil society organizations suffer from limited capacity in the face of such large problems as high levels of poverty and inequality. Such capacity limitations are attributed to their own lack of resources, given the number of demands that they face, and the negative effect of external environments which are not conducive to the creation or scaling up of civic participation. The Role of Government 49. Recent attention has turned to governments as important actors in promoting social capital by both providing an enabling environment for, and promoting more directly, social capital. Bratton (1988) and Evans (1996) suggest that it is the lack of an enabling environment at the macro level which generally provides the single most important obstacle-or facilitator-- to the scaling up of civic participation and ultimately to the promotion of social capital in developing countries. 50. The macro level factors necessary for promotion of social capital include: rule of law, transparent and effective governance, conflict mediation systems and general confidence in the capabilities of the state. These can tend to work against social capital if they are characterized by a general climate of distrust and insecurity which act as a disincentive to citizen participation (O'Donnell, 1994). As O'Donnell notes, these obstacles are common in countries which have recently adopted electoral democracies but which still lack democratic, responsive and accountable institutions, and where norrns and practices fail to provide new spaces for civic participation (1992, 1994, 1998). These characteristics are often absent in countries with histories of authoritarianism and clientalism. 51. While these limitations at the macro level present a constraint, they provide an opportunity for government action. Hadenius and Uggla (1996) provide an extensive discussion of the five stages of enabling social capital, presented in Box 4 , which can guide public governments to improve the environment for social capital as well as the promotion of social capital at the micro level. If Box 4 is interpreted simplistically as providing five set, sequential stages it may seem overly developmental. However, the characterization of state-society relations and specific actions taken by governments is helpful in better understanding their role in providing an enabling environment for the promotion of social capital. Government action can improve the climate by taking steps to ensure high levels of confidence and trust at the macro 11 level. They can also contribute to the development of social capital in communities by promoting spaces for civic action, channels for engagement,. and proactive support of independent association and collaboration. Thus, achieving high levels of social capital requires intense action at the micro level and complementary action at the macro level. Both governments and civil society organizations have important roles to play. Box 4: Moving from a Hostile State to a Benevolent State: Increasing the Odds of Social Development > First stage. The state, probably totalitarian or authoritarian, does not tolerate any autonomous organizations or independent civil activity. Moving from stage one requires that the state permit the formation of independent civic organizations. Threshold between stage one and stage two is de facto right to form autonomous organizations. > Second stage. The state accepts autonomous organizations but does not provide a space for themn. The state domninates provision of services. With reforms, however, the state's role is diminished. The existence of civil society organizations is tolerated and over time they gradually begin to perform traditional public roles. Threshold for moving to the next stage is that the state provides a space for independent activity. > Third stage. The state allows civic organizations to fil in gaps through service delivery but civic organizations are not consulted on sectoral policy; there is little collaboration between the state and these new organizations. Threshold for advancing to the next stage is that there are more favorable institutional structures. > Fourth stage. The state not only allows civic organizations to exist and to deliver services but it invites them to participate through favorable institutional structures. This requires providing channels for influence, arenas for interaction, and a facilitative legal-administrative framework. In terms of government action, this often entails effective decentralization which brings politics closer to citizens. Public sector reform is carried out in such a way that civic organizations are considered when making policies affecting them and their members, and that provides a legal framework which facilitates organization, resolves conflicts and enhances accountability to citizens. Threshold is that the state has active programs to support and develop civil society. > Fifth Stage. The country has a streamlined but effective state, with institutions that promote accountable governance. Citizens participate in resolving their own problems while interacting with the state on common ground. Modified from Hadenius and Uggla (1996) 12 III. A BALANCE SHEET ON SOCIAL CAPITAL IN ARGENTINA Taking Stock 52. This section seeks to take stock of the existing evidence on social capital in Argentina based on available past studies, surveys and other research. It examines macro levels of trust, as illustrated by general levels of confidence in institutions and existing evidence on participation levels, and provides a brief overview of the history of civic organizations in Argentina and their relations with the state. In doing so, it provides some initial hypothesis on the levels of social capital in Argentina and historical reasons for this. 53. Public institutions in Argentina are poorly regarded with the exception of educational institutions. A Gallup study (1997) found that the levels of confidence in public institutions, including national and local governments, was worryingly low. Table 2 shows that schools, churches and an array of civil society organizations draw more confidence than either municipal or national government. Table 2: Confidence that Certain Institutions can Resolve Social Problems High Confidence Low Confidence No Answer, Not Sure Public Schools 63% 32% 5% Catholic Church 58% 37% 5% State Universities 57% 29% 15% Nonprofit organizations 57% 33% 11% Civil associations 39% 38% 23% Small businesses 31_ % 59% 10% Municipal governments 26% 68% 6% Businesses 22% 69% 9% The army 22% 67% 12% Large businesses 19% 72% 9% National Government 19% 74% 8% Labor Unions 13% 78% 9% Political parties 11% 84% 5% Source: Estudio sobre Donaciones y Trabajo Voluntario, Gallup Institute, Argentina, May 1997 54. Within this context of low confidence in public institutions and despite a sudden growth in the number of civil society organizations, general levels of participation in Argentina are low compared to other countries. The World Values Survey, undertaken by the University of Michigan, periodically measures levels of interpersonal trust and participation in Argentina and 13 other countries. Table 3 confirms that Argentina has significantly lower participation and trust rates than other developing and developed countries. The numbers are the average of the levels measured in 1980 and 1990. Table 3: World Values Survey: Levels of Participation and Trust Country Trust Rate Participation Rate' Argentina 0.25 0.41 Brazil 0.07 0.87 Canada 0.51 1.40 China 0.60 0.98 Costa Rica 0.45 El Salvador 0.35 Guatemala 0.40 Honduras 0.50 India 0.34 Italy 0.32 0.50 Japan 0.42 0.42 Mexico 0.26 0.63 Nicaragua 0.32 Norway 0.63 1.50 Panama 0.43 Russia Fed. 0.38 1.03 Sweden 0.62 1.62 United States 0.48 1.66 55. A report undertaken by Facultad Latinoamericana de Ciencias Sociales (FLACSO) confirms these low participation rates and also provides a "snapshot" of civic organization's activities in Argentina, based on data from 3,678 organizations (1997). Only 20 percent of adults in cities are likely to participate. Women (22%) are more likely to participate than men (17%), and more highly educated individuals (29% with university education) are more likely to participate than those without higher education (17%). 56. The study also shows that 44 percent of civic organizations covered in the FLACSO study provide direct services - mainly in health, education and job training. Most organizations work with children and youth (47%), 12 percent serve senior citizens, 11 percent work with rural 6 workers and only 8 percent focus on the handicapped. The FLACSO data also capture important regional dffferences in the density of organizations (see Table 4). Most organizations are found in the largest urban areas: Buenos Aires, C6rdoba, Santa Fe, and the Federal District. 5 Participation rates are calculated as the average number of organizations in which citizens participate and can therefore be more than one. These results are the average values of the 1980 and 1990 World Values Study. 6 It should be noted that this study only presents simple correlations and therefore this data cannot be interpreted as proving causality 14 Table 4: Non-Governmental Organizations by Province Metropolitan Provinces Number of Peripheral Provinces Number of Organizations Organizations Buenos Aires 604 Chaco 131 Capital Federal 489 Jujuy 110 C6rdoba 227 La Pampa 161 Mendoza 620 Rio Negro 121 Santa Fe 225 Salta 203 Subtotal 2165 Tucuman 174 Subtotal 900 Remaining 13 633 provinces A Historical Perspective 57. A brief overview of the history of civic organizations and their relations with the Argentine state can shed light on the potential causes of high levels of public distrust and low levels of participation in Argentina. As shown in the following summary, Box 5, civic organizations in Argentina have been deeply affected by sharp changes in the country's political and economic circumstances, a blurred delineation between the public and private spheres, and a tradition of clientalistic relations whereby civic organizations are formed by the state and often ultimately co-opted by them (Cavarozzi and Palermo, 1995). 58. Civil society organizations existed in Argentina before independence. Thompson (1995) and Campatella and Gonzalez Bomball (1999) trace the origins of organized civil society back to colonial times (see Box 5). While presenting such a wealth of material in such a box risks oversimplifying certain periods or events, Box 5 provides a useful and quick overview of the history of civil society organizations for the unfamiliar reader. During the colonial period, the Church dominated "acts of charity." Despite the asistencial nature, these faith-based organizations contributed much to Colonial Argentina in the form of hospitals and orphanages. The well-known Hospital de las Mujeres and the Casa de Nifnos Expositos were founded during this period. 15 Box S. Chronological Synthesis Period Characteristics Organizations of the Third Sector Colonial Period * Strong influence of the Catholic * Charitable organizations of the Church (Until 1816) Church o Religious orders (Jesuit, Franciscan, etc.) o Brotherhoods (Brotherhood of the Santa Caridad) Independence Period * Absence of a central state * Dissolution of the Broterhood of Santa Caridad (18161880) * Secularization of the sector * Creation of the Sociedad de la Beneficencia by the state which is: o Dependent on the state o With autonomy in terms of funds raising from private sector o L'ed by ladies of high society Consolidation Until 1912 Of the State and * Consolidation of the Central State * Consolidation of the Sociedad de la Beneficencia Emergence of Civil * Unification of a domestic market * Creation of numerous charitable organizations which: Society around an agro export model o Tied to the Church, and (18801945) * Superposition of the public and the o Led by the high society with contributions from the state private sector * Organizations of immigrants and the working class (mutual, clubs, * Wave of foreign immigrants to schools, etc.) Argentina 1912 -1930 * Broadening of the political system * Development of trade unions, associations of mutual aid, * Regulation by the state of Civil neighborhood associations society organizations * Implement popular development associations, libraries, etc. 1930- 1945 * aD6cada infame * Neighborhood associations and self set up organizations along with o Coup d'etat and restoration of a organizations of the Catholic Church and the ladies of high society fraudulent and restrictive political system Populism and Social . Rise of Peronismo . Growth of trade unions Rights (1946 - 1955) * Strong relationship between a * State assumes social protection functions, previously carried out by regulatory state and civil society private foundations o Extension of social rights * Dissolution of the Sociedad de la Beneficencia o Rupture of tie between the State and philanthropy exerted by high classes o Creation of the foundation Eva Peron New neighborhood associations - more political in character Proscription and * The military coup against the * Organizations of the third sector either coopted by government or by Social Conflict Peronist government the opposition (1955-1983) * Government initiates period * Intervention and dissolution of the Foundation Eva Per6n characterized by * Growth of the power of the trade unions o Institutional instability * Creation of centers and cultural foundations as a result of the exodus of o Constitutional changes academics from Public Universities o Politicization * Growth of organization of human rights o General political violence Transition and * Transition to democracy * Consolidation of organizations of the third sector which begin to Consolidation of the o Greater autonomy of civil society experiment with autonomy from the state Democracy (Part of * Redefinition of the role of the state: * CSOs occupy spaces that once belonged to the state 1983) pulls out from traditional functions * Development of philanthropic organizations tied to companies: birth of * Consolidation of an economic model social responsible businessmen of "Free Market" 59. After Independence, the new government strove to limit the role of the Catholic church in charitable activities. Despite the government's zeal, its own capacity to deliver social assistance was limited, and so in 1823, in an attempt to provide social assistance through a private, secular source, the Govermment of Buenos Aires brought together the wives of Argentine high society to form the Sociedad de Beneficiencia. The Sociedad took on the character of a private foundation. Assistance was awarded on a discretionary basis and'was well intentioned (Thompson, 1995). In 1912, the National Budget stated that 69 percent of social assistance was covered by private acts of the Beneficiencia and other private non-profit organizations (Canipetella y Gonzalez Bombal, 1999). 16 60. After the World War I, Argentina experienced a great influx of immigrants from Europe who, particularly Italians, soon formed a new array of civic organizations. During this period, the mutual support networks grew rapidly, concentrating in Buenos Aires but with representation in Santa Fe (12%), Entre Rfos (5.8%), C6rdoba (5.4%), Corrientes (3.8%) and Tucuman (1.0%) (Campetella y Gonzalez Bomball, 1999). These self-help networks developed into an impressive number of rural cooperatives and urban neighborhood associations.7 A historical summary of the neighborhood associations provides interesting insights into the traditional role of civic organizations (see Box 6). Box 6. The Urban Movement: Neighborhood Associations in Argentina Neighborhood associations in Argentina refer to a set of organizations which are established in an attempt to meet specific needs of their members: urban residents. Neighborhood associations were born between World War I and World War II to meet the collective needs of new immigrants arriving from Europe. This wave of immigration, along with the industrialization of Argentina, led to the rapid growth of urban sprawl. Neighborhood associations became effective organizations for ensuring that the transport and infrastructure needs of new communities were met - originally through self-help, and eventually through negotiations with the government (Campetella and Gonzalez Bombal, 1999). The Neighborhood Associations represent the first set of civic groups to form around issues of location and a new urban cultural identity, rather than race or religion (Silva, 1995). A number of clubs and sub-groups formed under these umbrella associations, including the well-known "popular libraries" and other initiatives which attempted to both meet the needs of Argentines that settled in areas which were yet to be reached by public services, as well as promote a collective identity among the new urban popular classes. While some of the Neighborhood Associations represented a new growth in participation, they soon became the object of hierarchical and bilateral relations with the state, which endeavored to control or redefine them through legislaton or informal practices (Cavarozzi and Palermo, 1995). Thus, the Neighborhood Associations were formed and functioned under a state-centric model of participation. The associations were used solely as a channel for the community to voice demands before the state and then to deliver public goods back to the population which, in return would provide its labor free. While the relationship between existing neighborhood associations and the Peronist government was strained, the state continued to be the center of the group's focus, as their non-Peronist leaders of the associations used their group to pressure the government into providing short-term benefits for their members. A consequence of these vertical ties was the competition for resources between individual organizations and the disincentive to create broader networks of neighborhood associations. In addition, these groups became "hired-guns" for the delivery of public works without ever leveraging a role in the formulation of public policies (Cavarozzi and Palermo, 1995). Following the military dictatorship, there was some hope that the Neighborhood Associations would grow and further strengthen their independent capacity and self-management practices that had developed out of necessity under military rule. However, these groups were unable to shake their clientalistic practices of old and, with the reduction of public works budgets under economic reforms, these associations also ceased to perform their role as effective channels for community needs. Instead, the Neighborhood Associations became political platforms for local politicians who used their members to secure votes for local party leaders known as punteros. In recent years, the number of neighborhood associations have increased but their role is increasingly political, not civic, and their membership has dwindled. As a result, the activism and participation around these associations is declining and the remaining proto-organizations have become "essential elements of their leaders' ambitions" rather than a form of civic engagement. 61. The early part of the twentieth century also saw the birth of the early form of trade unions. 'Workers circles" or guilds were originally promoted by the social theology movement within the Catholic Church in an attempt to help workers and their families meet their needs 7For an interesting case study on rural cooperatives, see Roberto Benencia (1992). 17 through self-help initiatives. In 1916 the first Congress of Organizations of Self-Help and Social Protection was held, and a regulatory framework for Organizations of Self-Help was produced. This marked the first attempt by the state to provide a regulatory framework for civic participation. 62. Under the military government, which assumed control in 1930, continuing efforts were made to provide order and regulate civic organizations. The military government attempted to take stock of the number of private organizations by initiating a national registry for civic organizations (Thompson, 1995). In 1937, they also made a first attempt to define non profit status under the new tax law, and exemptions were provided to many cooperatives and neighborhood associations. Thus, the first half of the 20th century can be characterized as a time of growth for secular civic organizations, and an attempt by the state to provide some order. 63. The re-ordering and demarcation of the civil society was subject to another change under the government of Juan Per6n. The Peronistas came to power in 1945 and Argentina underwent further social, political and economic changes - including a new Constitution in 1949. During the administration of the Peronistas, the role of government grew and the national systems of social security and health assumed - albeit in a more comprehensive fashion - the role of the self-help associations (Cavarozzi and Palermo, 1995). At the same time, civic organizations, which had sprung up to meet the immediate needs of workers or immigrants, now began to assume a more political nature. 64. Three types of civic organizations illustrate this process of politicization. Trade unions are the most well-known example of the new politicization and the blurring of roles between the public and private spheres. The trade union movement played an important role in Juan Per6n's rise to power and, as a result, was a close ally, proving instrumental in the design of the new social security and health system. 65. This blurring of distinctions between the public and the private, as well as the politicization of civic duties of private associations, was also evident in the Fundaci6n Eva Peron. The Eva Peron Foundation was formed to attend to the needs of non-workers and their families. The Foundation shared many characteristics with the first Argentine foundation, the Sociedad de Beneficiencia. Like the Beneficiencia, the Fundaci6n Eva Peron was designed based on the feminization of the provision of social assistance. And, like the Beneficiencia, it was established by the government but acted as a private foundation with generous public funding. However, as Campetella y Gonzalez Bombal note, the Fundacio'n Eva Per6n also represented a break with traditional networks of assistance (1999). Fundaci6n Eva Per6n promoted a network of popular women's neighborhood associations. Furthermore, the rationale behind the new Foundation was one of social rights and justice for all citizens rather than just one of charity and assistance for the needy (Per6n, 1973). 66. The third example of a hybrid of civic-political organizations is provided by the new type of neighborhood association: the unidades bdsicas. These local associations shared many characteristics with the early neighborhood associations. Like their predecessors, the unidades bdsicas were based on the concept of belonging to a community and sharing an identity and culture with neighbors. Yet these neighborhood association were more political in their purpose than their predecessors and, in effect, sought to serve as local arms of the Peronist party. 18 67. The military coup in 1955 provided Argentine society with yet another abrupt change and challenge in terms of social organization, and started a process of uncertainty which continued for the next 25 years. Civic organizations tended to fall along two lines: those that resisted the military dictatorship or groups that supported it. In addition to the repression of civic action and the divisions caused under military rule, this period saw two additional developments. The first was the development of new cultural and scientific organizations, and the strengthening of local neighborhood initiative, as academics and intellectuals expelled from public universities for political reasons joined their ranks. The second reflects the agenda of international funding for organizations which sought to defend human rights and victims of the military dictatorship. This period it generally looked upon as one where civic organization was repressed and democratic traditions suspended (Campetella and Gonzalez Bombal, 1999). 68. The political democratization which started in 1983, and the economic reforms which were subsequently adopted, shaped the most recent period of the history of civic organizations in Argentina (Thompson, 1995). Civic organizations played a central role in the democratic transition and the period is characterized by some as a "resurrection of civil society" (Campatella and Gonzalez Bombal, 1999). Imniediately after the ousting of the military government, there was a growth in the number and type of organizations. Some sought justice for disappeared relatives and others championed new issues such as the environment, women's rights and poverty reduction. At What Stage is Social Capital Now? 69. Despite the recent growth in the number or organizations, some authors hypothesized that this would have little impact on the aggregate levels of participation, which would remain low. Three reasons have been cited for this. The first suggests that with the changing role of the state, civil society organizations have struggled to redefine their relations with the public sector and at present are still in a period of redefinition of roles and responsibilities (Beard, 2000; Cavarozzi and Palermo, 1995). The second suggests that the tradition of clientalistic relations, which have characterized many state-civic organization relations, has resulted in broad skepticism among many citizens. Based on fieldwork undertaken at the end of the 1980s, Cavarozzi and Palermo (1995) state that participation is seen as an "exercise in futility" which they attribute to the "clientalistic leadership which diminished the scope and salience of participation", especially in low-income communities in urban centers. The third suggests that, while the number of organizations has grown, their capacity remains somewhat limited. Filmus et al (1997) suggest that this is due to: (i) lack of information about development projects and funding sources; (ii) relationships with different levels of the state that are burdened by clientalism and excessive bureaucracy; and (iii) weak organizational capacity and continued dependency on the state. Thompson also adds that the competition among organizations for funding and weak institutional development present potential weaknesses within the sector (1995). 70. In recent years, there has been an attempt by the public sector to engage civic organizations in collaborative efforts. Thompson cites examples of CSO collaboration in public sector rural development projects, labor training of the unemployed and community capacity building projects. In addition, a National Center for Civic Organizations (CENOC) has been established in the Secretary of Social Development to provide data and information on civic 19 organizations. These initiatives seem to be limited to collaboration around service delivery in areas where civic organizations may have comparative advantages. There is little evidence of a broader effort by the public sector to systematically engage civic organizations in the policy formulation process or to promote the independent capacity of civic organizations through their own projects. 71. Thus, it would seem that Argentina is at Stage 3 in the state-civil society activity presented earlier in Box 4. The situation could be characterized by the following: * Sharp social, political and economic upheavals seem to have been a constant feature of Argentina's history, with dramatic changes in social institutions. * The democratization in the 1980s along with the macroeconomic reforms in the 1990s, have once again transformed the nature of state-society relations, but this transition has been incomplete. * In particular, the capacity of public institutions to respond to the new demands and ways of operating in a transformed economic and social model has not kept pace and has led to dissatisfaction. * There is clearly space for civil society organizations in Argentina, and many have had a very positive impact. Numerous organizations have developed, especially in the democratic period, but capacity may still be limited. * Yet, the tradition of clientalistic relations between the state and civic organizations, along with low levels of confidence in institutions, levels of insecurity, and charges of corruption of public officials, provide a difficult climate for the promotion of social capital. 20 IV. A NEW QUANTITATIVE ANALYSIS Description of the Tool 72. A social capital survey was specifically designed for this study. The survey draws on the Social Capital Assessment Tool (SCAT) which was designed by the World Bank8 based on a comprehensive review of 26 social capital surveys from 15 countries. The survey used for this report also incorporates significant modifications. The modifications are designed to: (i) better tailor the questionnaire to the Argentine context; (ii) base trust questions on a series of hypothetical questions about actual behavior at the individual level rather than impressions of trust at the community level; and (iii) shed light on the links between the environment for social capital and micro levels of social capital. 73. The Argentine survey is divided into four sections. Section 1 contains 18 questions on the socio-economic status of the respondent, including similar information to that used in the National Census. Section 2 contains 29 questions relating to participation in formal organizations and forms of spontaneous collective action. This section is designed to assess: (i) levels and reasons for participation; (ii) characteristics of participants; (iii) types of organizations that Argentines participate in, in terms of homogeneity of membership, bonding or linking groups, and internal decision-making. Section 3 is designed to measure collective action during times of crisis and during "normal" times.9 The survey captures collective responses to both idiosyncratic and systemic shocks, as well as collective action as a strategy for influencing decisions and leveraging public resources during "normal" times. Section 4 captures levels of trust and solidarity. Based on a series of hypothetical questions, we assess the levels of inter-personal trust of respondents. Section 4 also contains six questions relating to public perceptions of institutions and general levels of trust beyond the community to assess the relationship between respondents who participate more or have higher levels of interpersonal trust at the micro level, and broader levels of confidence and social trust. 74. The survey was field tested in Argentina during May 2000. It was then applied to a nationally and regionally. representative sample of households during the months of May-June 2000. The sample was also representative of rural and urban communities alike. The number of 8 See Krishna and Shrader (1999) 9 Collective action is defined as a spontaneous, self-organizing process whereby a social group (or group of individuals) attempts to influence decisions or access new resources (economic, political and cultural), Unlike participation, collective action does not imply the presence of permanent structure or formal organization with legal identity. For a fuller discussion of this term see Sheldon Annis (1988), John Friedman (1992) and Charles Tilly (1981). 21 responding households totaled 2,235. The rural communities sample is based on small municipalities of 5,000 habitants or less. Consistency and Improving the Measurement 75. The questions on both participation and trust in the Argentine survey offer an improvement on past surveys, such as the General Social Surveys (GSS) of the United States and the World Values Survey (WVS) sponsored by the University of Michigan. The questions used for participation aim to measure not only organizational density and the type of organization in which respondents participate, but also to gauge a series of internal characteristics of the organization. Thus respondents were first asked to identify the organizations or groups in which they participate. Possible responses included 31 different types of civic organization - some formal, some not, some with specific welfare goals and others with broader networking goals. Respondents were then asked a series of questions about the internal workings of the primary organization, including homogeneity, decision-making, and the purpose of the organization. These questions help to measure more than the act of participation and shed some light on the expenience of participation, providing helpful insights into why some groups seem to promote higher levels of trust than others. 76. In the same way, the Argentine survey tries to overcome the shortcomings of trust questions found in previous surveys. Glaeser et al (2000) provide an extensive discussion of the difficulties in measuring trust rather than trustworthiness. The problem with trust questions lies in their ambiguous phrasing, their leading nature, and the difficulty in ascertaining whether individuals are responding for themselves or rather according to their perceptions of community- wide levels of trust.10 To overcome this problem, the following hypothetical was inserted which helps to limit the respondents' response to a realistic situation at the individual level: "If you had to leave suddenly, whom could you trust to take care of delicate matters such as your house, your children or a dependant adult?" If the respondent cited any one beyond the family unit (but possibly including the family) then they were considered to trust others1'. The Survey Results 77. The results are presented in five sections. The first section presents the findings of general participation rates, levels of participation by population and region, and findings on group leadership and internal organizational characteristics. The second section presents findings on collective responses to shocks, and collective action as a more permanent civic strategy. The third section presents general and regional levels of inter-personal trust, characteristics of those that trust, and interpersonal trust with internal organizational characteristics. The fourth section presents baselines for both participation and inter-personal trust. Finally, in the fifth section, the results on the relationship between confidence in public institutions and broader levels of trust with micro levels of interpersonal trust and participation are presented. 10 Thus, when the respondent is asked whether they agree with the statement, "the majority of people in this community are basically honest and you can trust them," it is not clear whether they are referring to his/her own levels of trustworthiness or to community-wide levels of trust " Responses were not limited by the questionnaire. 22 78. The results are based on two exercises: cross tabs and regressions using a probit model. The results hold even after the inclusion of regional, provincial, or cluster dummies, and are based on regressions in which the standard errors have been corrected for clustering. A. Levels, Types and Determinants of Participation 79. Levels of Participation. The participation rate in Argentina is generally low. 18.3 percent of all respondents said that they participated in at least one organization. A smaller proportion said that they participated in two organizations (2.2%) and only 0.2 percent said they participated in three groups. These findings are similar to the Gallup Institute study cited earlier. That study found that 20 percent of adults in major cities participate in civic organizations (1997). Participation is low despite the growth or organizational density claimed by recent studies of "Argentina's Third Sector" (Johns Hopkins, CEDES 2000). 80. This level of participation is significantly lower than the membership density measured in the World Values Survey (WVS) in Argentina for 1980 and 1991 However, this difference should be treated with some caution as the methodological differences between the two surveys in terms of participation are significant. Since the WVS question is somewhat more leading and provides a list of organizations, respondents are more likely to respond positively. The same WVS data did find Argentina to be the country with the lowest level of membership density in a sample of 37 developing countries. Our findings confirm that the general participation rate in civic organizations in Argentina is indeed low. 81. Characteristics of Participants. Table 5 shows the characteristics of those Argentines who are more likely to participate: women, household heads, those that are married, have marginally more education, are slightly older and are employed are more likely to participate. Table 5: Characteristics of Individuals Who Participate Participants Non-Participants Demographics Male .4393 ** .4876 Household Head .3498 .2942 Married .5451 .4374 Years of Education 10.79 * 10.28 Age 40.60 *** 37.85 Economic characteristics Employed .4955 n.s. .49476 Unemployed .0805 ** .1313 Notes: Estimated means, taking into account the clustered and weighted nature of the data. Levels of significance for an F-test for difference in the means for participants and non-participants are reported. * Difference is significant at the 10% level; ** difference is significant at the 5% level; *** difference is significant at the 1% level; n.s.: difference is not significant. 23 82. The picture for participation and income is somewhat more complex. A simple correlation between the probability of membership in an organization and income suggests that the income of members of groups tends to be somewhat higher than that of non-members. However, the relationship between the two variables-membership and income-is not significant either in the larger sample (which includes information on respondents, as well as other household members) or in the smaller sample (which includes information on respondents only). 83. The results from a non-parametric regression'2, shown in Figure 2, are similar: the estimated probability of membership goes up with log income, although it is not clear whether the relationship is significant.'3 Figure 2 also suggests that there may be some curvature in the relationship between log income and the probability of membership, so that the poorest and richest individuals have a higher probability of membership than those in the middle of the income distribution. Figure 2: Probability of membership and log per capita income 12 Regressions are fan regressions, see Fan, Jianqing (1992) 3Bootstrapped standard errors would have to be computed to see whether the relationship is significantly different from zero, and this was not done. 24 .25 - n .2 E E 11 5 .05 - 0 0. 0 .05 2 3 4 5 6 7 Log of per capita income 84. Describing the relationship between log income and membership in organizations is difficult because the direction of the causality is unclear as membership could be a function of income: for example, richer people have more time at their disposal, and may use that time to join organizations, or poorer people join organizations as a means of protecting themselves from shocks because they have fewer other means (such as savings or credit). 85. But income itself could be partly determined by membership, for example, if the networks developed in organizations lead to higher earnings. It is therefore not clear whether income "belongs" in a membership equation, or whether membership "belongs" in an income equation. In a simple multivariate regression framework, simultaneity of this sort can coefficient on income in a membership equation, as well as the coefficient on membership in an income equation.'4 86. Because of these problems associated with the possible endogeneity of income in a membership equation, four sets of membership regressions are reported in Table 6. The first two columns ((i) and (ii)) are based on the larger sample, while the next two columns ((iii) and (iv)) are based on the smaller sample which is limited to respondents only. In each case, we limit the sample to those aged 16 and over, and include a vector of cluster-level dummy variables to control for unobserved cluster-level heterogeneity. All estimations are done by probit. The Table reports the estimated probable derivatives at the mean and standard errors are corrected for clustering. We present regressions which do not include income (columns (i) and (iii)), and those which include parameters for the log of income, and the squared log of income (columns (ii) and (iv)). 14 One solution to this problem is to use instrumental variables as an estimation technique. This would involve purging (say) the coefficient on income in the membership equation with a variable, the instrument, which is correlated with income but uncorrelated with the error term in the membership equation. A Haussman test could then be used to test whether income was endogenous in the membership equation to begin with. Unfortunately, no suitable valid instrument was available. 25 Table 6: Determinants of Membership, Various Specifications (i) (1l) (W) .(IV) Age .0037(2.42)** .0034(2.22)** .0084(3.24)*** .0078(3.02)*** Age squared (x100) -.0030(-1.86)* -.0029(-1.85)* -.0069(-2.48)** -.0064(-2.32)** Years of education .0062(3.82)*** .0041(2.94)*** .0099(3.13)*** .0082(2.69)*** Male -.0320(-4.06)*** -.0332(-4.10)*** -.0294(-1.87)*** -.0318(-1.99)*** Married .0209(1.76)* .0224(1.87)* .0001(0.01) .0038(0.22) Household head .0277(2.83)*** .0267(2.78)*** .0059(0.35) .0041 (0.24) House owner .0232(-1.60) -;0238(-1.66) 0187(-0.93) 0181(-0.92) Log income - -.0468(-1.19) 1231(-1.88)* Log income squared - .6654(1.77)* - 1.3600(l.99)** (xlOO) Observed probability .142 .142 .202 .202 of membership Predicted probability .132 .131 .190 .190 of membership Pseudo R-Squared .0475 .0503 .0526 .0548 Number of 6,249 6,249 2,102 2,102 observations Notes: Dependent variable is the probability of membership. Z-statistics corrected for heteroskedasticity and clustering are reported in parentheses. * Significant at the 10% level; ** significant at the 5% level; *** significant at the 1% level. 87. The regressions confirm that age, age squared, years of completed schooling, gender, marital status, and household headship are all signih'cantly associated with the probability of membership. Specifically, based on the specifications in column (i), the probability of membership increases monotonically with age, albeit at a decreasing rate; every additional year of schooling increases the probability of being a member in an organization by about 0.6 percent; while being a woman, married, and a household head increase the probability of membership by 3.2 percent, 2.1 percent, and 2.8 percent, respectively. Controlling for these other variables, there may also be a U-shaped relationship between membership and income, so that individuals with the lowest and highest incomes are more likely to be members in organizations than those in the middle range of the income distribution. However, the results on income presuppose that income is truly exogenous in the membership equation, and do not hold for the larger sample which includes both respondents and other household members (column (ii)). 88. Regional Levels of Participation. Table 7 presents participation rates by region and finds that there are significant differences. It is clear that there are important regional differences 26 in participation with the highest level being in the South, followed by Cuyo and the Central region. Interestingly, in the Metropolitan region where organizations are closest to the central government and decision-making, participation rates are the lowest. Table 7. Participation Rates by Regions (%) Region 1: Region 2: Region 3: Region 5: Region 6: I MetropoL I Central I Northeast Cuyo Northwest South 14.5 21.3 20.9 22.4 19.4 26.3 Source: Calculatons based on survey data compiled by Sofres-lbope, Argentina. 89. Reasons for Participating. Table 8 presents the reasons offered by respondents for participating in an organization. The most common reasons cited by respondents who participate include: (i) feeling useful, (ii) doing something for others, (iii) spiritual development, (iv) to meet people and not feel so alone, (v) to share things which one believes in, and (vi) learning new skills. Feeling useful is the most frequently cited reason for participating and it is even more important for the unemployed. Married and unmarried women are more likely to participate to learn new skills then their male counterparts. Interestingly, participation is rarely seen as providing a monetary benefit or a tangible good such as an insurance mechanism. Table 8: Reasons Offered By Respondents for Participating in an Organization Reason Total Demographic C aracteristic Men Women Employed Unemployed Inactive To feel useful 1 6 23 9.7 16.4 13.4 I To do something for others 1 2 22 10.5 12.7 11.4 1 2 Spiritual development 1 2 39 11.0 10.9 9.6 0 4 4 . _ Meet people, not be alone 8 2 19 8.1 10.1 7.6 _ _ _ _ _ _ _ _ _ _ _ _ _ ~~~0 7 _ _ _ _ _ _ Share one's beliefs 9 7 12 10.1 9.5 8.6 New skills, valuable training 6 1 32 5.4 8.1 7.9 _ _ _ _ _ _ _ _ _ _ _ _ _ ~~~7 4 _ _ __ _ Sense of protection against problems 3 3 9 3 7.7 3.8 7 _ Others 6 4 12 57.8 75.4 69.9 1 0 4 Source: Cunningham (forthcoming) 90. Leadership of Organizations. Table 9 shows that, although men are less likely to participate than women, those men that do participate are more likely to be the leaders of the organizations. This is the only significant characteristic in determining leadership. Marital status, age, years of education, income and employment status are not significant in determining leadership, although this is likely to be due to the small sample size. 27 Table 9: Characteristics of Individuals Who are Leaders Leaders Other Partidpants Demographics Male .6428 * .4329 Household Head .5653 n.s. .3430 Married .5607 n.s. .5446 Years of Education 11.53 n.s. 10.76 Age 42.38 n.s. 40.55 Economic characteristics Income per capita 182.38 n.s. 216.01 Employed .7282 *** .4882 Unemployed .0242 ** .0823 Home-owner .8445 n.s. .7748 Notes: Estimated mneans, taking into account the clustered and weighted nature of the data. Levels of significance for an F-test for difference in the means for volunteers and non-volunteers are reported. * Difference is significant at the 10% level; ** difference is significant at the 5% level; *** difference is significant at the 1% level; n.s.: difference is not significant. 91. Types of Organizations. There are four ways to assess the type of organization in which Argentines participate. First, we look at the activities of the organization. Second, we turn to the membership of the organization to assess the level of homogeneity or heterogeneity of participants. Third, we assess whether the organization is one which bridges outside the organization or whether its focus is primarily a bonding one. Finally, we look at the internal decision-making processes within organizations to better understand the internal functioning of organizations in which Argentines participate. 92. Activities of Organizations. Figure 3 presents participation by activities. Religious organizations, which are identified by respondents as a type of organization, represent the single most popular type of organization. Both Catholic and non-Catholic churches and their affiliated groups, draw the largest numbers of members with 47% of respondents who participate identifying the church or a religious group as their primary organization. While the Catholic church is still the most popular, evangelical groups and Jehova's Witnesses also figure prominently. Of the secular groups cited by the respondents who say they participate, a broad array of self help groups are the next most important (28%), followed by sports clubs (12%).15 The self help groups include community based organizations, parents teachers association and farmers unions. Men are more likely to participate in sports clubs and women in church groups. Political parties and trade unions figure as less important (3%). 15 It should be noted that sports clubs traditionally play an impoitant social role among immigrant and middle class populations ( see Campatella and Gonzalez Bombal, 1999). 28 Figure 3: Types of Organizations in Which Argentines Participate Other Political and Trade 10% Unions 3% Sports 12% \ iReliglous 47% 28% Source: Calculations based on survey data compiled by Sofirs-lbope. 93. Homogeneity and Heterogeneity of Membership Twelve questions were asked to respondents about the other participants in their organization.16 Then, two measures of homogeneity were constructed. In one measure, respondents were deemed to participate in homogeneous organizations if they answered "yes" to at least four attributes shared by all members. In the second measure, respondents were considered to participate in homogeneous organizations if they answered "yes" to at least seven attributes shared by all participants.17 Regardless of the measure used, females are more likely to join homogenous organizations than males, and the poor are more likely to participate in homogenous organizations. Thus, the poor are more likely to participate in organizations where other members share similar socio- economic backgrounds. This mnay reduce their capacity to leverage knowledge and resources and influence decision-makers when seeking solutions to common problems, as it limits their contact to individuals from similar walks of life. The rich, in comparison, are more likely to belong to heterogeneous organizations. This enables them to renew contacts and leverage resources outside of their traditional circle of contacts and influence. Figure 4 presents the average monthly income (AR$) of members of homogeneous and heterogeneous organizations. Figure 4: Average Monthly Income of Members of Homogeneous and Heterogeneous Organizations 300 $253.98 a 250 $192.36 ~200- 16 The twelve quesi 0 on, sex, political beliefs, occupation, age, e 8 hical origins of members, nationality of mem . 50 7 The results reporte O Heterogeneous Homogeneous Source: Calculations based on survey data compiled by Sofres-lbope, Argentina. Note: Differences in levels of income and homogeneity are significant at the 5% level, or higher. 94. Bonding and Linking Organizations. To assess whether the organizations perform primarily bonding or linking functions, we asked participants to identify the type of activities that the organization performns. Respondents identified a series of "services and uses". Bonding functions are given to those services that are offered within the group such as self-help and mutual support activities. Those uses or services which require connections across groups and institutions are determined to be linking organizations. When we cross this with income levels, we find that the poor are less likely than the rich to participate in linking organizations and are more likely to participate in organizations of self-helIp and mutual support (see Figure 5). The poor, then, are more likely to belong to organizations iwhich perform social protection functions and less likely to belong to organizations which might serve as conduits for long-term transformation. The rich are more likely to belong to organizations which offer new contacts and opportunities. The poor thus participate to "get by" and satisfy their basic needs while the non- poor are able to participate in organizations which provide them with new opportunities to "get ahead." Figure 5: Average Monthly Income of Members of Linking and Bonding Organizations 300 $249.00 ! 250- o200 -$187.00 & 150- 8 100 -I 8 50- 0 Linking Bonding Source: Calculations based on survey data compiled by Sofres-lbope, Argentina. Note: Differences in levels of income and linking and bonding are significant at the 5% confidence level, or higher. 95. Decision-Making in Organizations. Table 10 presents participation by different types of decision-making within the organization. These characteristics are used to assess whether organizations are inclusive of their members and build the capacity of their members in institutional matters such as leadership, decision-making and fund management, or, whether they are managed by a few individuals in a hierarchical manner. Of the organizations cited, 79 percent are felt to involve their members in general decision-making, 70 percent engage members in selecting their leaders, 60 percent have male leaders, and 58 percent involve their members in fund management. Table 10: Participation Rates by Internal Characteristics of Organization Type of Organization Participation Rate (% of those that participate) Participatory Leadership Selection 70% Participatory Fund Management 58% 30 Type of Organization Participation Rate (% of those that participate) Participatory Decision-Making 79% Male Leadership. 60% Sou: Calculations based on survey data compiled by Sofres-roope, Argentina. 96. However, when these organizational characteristics are crossed with monthly income, we find that those with higher income are more likely to participate in the selection of their leaders (Figure 6) and in the management of the organization's funds (Figure 7). The poor's organizations tend to be more hierarchical, and thus they are less likely to be engaged as active participants in decision-making than the organizations of the well-off. This is likely to limit their opportunity to learn from the experience of participation and acquire new skills in institutional matters. Figure 6: Average Monthly Income of Members Who Participate in Selection of Leaders a3250 $252.00 250- co 200 200 8so- =t~~~~~~$530 ~*150 100 8 50 0 Participate In Selection of Do Not Participate in Leader Selection of Leaders Source: Calculations based on survey data compiled by Sofreslbope, Argentina. Note: Differences in levels of income and participation in selection of leaders are significant at the 5% level, or higher. Figure 7: Average Monthly Income of Members Who participate in Fund Management a 250 -200- o 100 50 - 8 50 - Participate in Fund Do Not Participate in Management Fund Management Surce: Calculations based on survey data compiled by Sofres-lbope, Argentina. Note: Differences in levels of income and participation in selection of leaders are significant at the 5% level, or higher. 31 97. In the trust section, we will also find that organizations that involve their participants in the selection of their leaders and in the management of their finances tend to promote higher levels of interpersonal trust among their members. B. Collective Reponses to Shocks 98. In this section we look at collective responses to shocks and collective organization as a strategy to engage public officials.18 This type of organization is often less structured than the participation assessed in the previous section, and frequently used as a spontaneous measure to meet specific needs. 99. Friends and Families. Argentines are most likely to turn to their family and friends when faced with any type of shock. Whether the shock is idiosyncratic, such as a job loss, or systemic, such as a natural disaster or widespread lay-offs, Argentines are most likely to turn to a close circle of family and friends around them.19 Table 11 presents the sources of support when faced with an idiosyncratic shock by gender, and household structure. Sixty seven percent feel that they would turn to their family and a significant portion feel that there is no-one to turn to (15.11%). Only 3 percent would turn to their neighbors and 8 percent turn to their friends. Men are more likely to feel that they have no-one to turn to but are twice as likely to depend on their friends. Women are more likely to turn to their families for support. Single mothers, perhaps surprisingly, are no different than two-headed households, viewing family and friends as the most important sources of support. It is important to note that this finding should not necessarily be interpreted as indicative of a lack of social capital in Argentina. Rather, it should be viewed in the context of the important role that family and extensive family in Argentina has traditionally played in providing relatives with support. Clearly the family continues to be an important source of support in times of crisis. Table 11: Assistance in the case of job loss (idiosyncratic) (%) 9051m " 15.11 16.49 13.80 9.73 15.14 F X Il_ ~~~~~67.10 64.97 _ 69.14 68.47 69.8 1 ~~3.09 2.75 3.42 6.68 1.54 0 : _ ~~~~~~8.47 10.39 -6.63 8.74 7.22 _ ~~~~~~~1.72 1.60 1.84 3.9 1.6 :i*overnmet _0.92 0.53 1.29 0.00 1.04 Rj~I~w~Z = _ 0.75 0.75 0.75 1.36 0.8s_ _ X giim!r 1,488 692 796 III 1,110 Source: Cunningham (forthconing) 100. Support during Unemployment. The responses of the unemployed in Table 12, when asked who they would turn to when faced with unemployment are particularly interesting since 18 See footnote 11 for definition and discussion of collective action. 9 For this section, only responses from household heads or their spouses were considered as they were felt to be in a better position to comment on household coping strategies. 32 they are based on real experience. The unemployed are less likely to turn to family and friends, and a greater proportion of them feel that they have no-one to turn to. Both they and their employed counterparts do not feel that the church, government, trade unions or civic organizations play a large role in providing assistance during these periods. The poor and informal sector workers also feel less optimistic that they can turn to anyone and are also less likely to turn to their families than their wealthier or formal sector counterparts. This table also shows that there is surprisingly little difference among income groups albeit the poor and unemployed would turn slightly more often to the church, government, and unions, demonstrating the self-reliance of the rich. Table 12: Assistance in case of job loss by employment and unemployment groups X~~~~~~~~~~~~~~~~~~ _ E,uiB i 15.11 18.63 13.73 19.49 14.55 13.29 12.34 14.84 X~~y . 67.1 65.86 67.58 61.65 71.35 68.06 71.47 67.83 3.09 4.00 2.74 4.68 2.79 1.35 2.69 3.03 8.47 3.15 10.56 7.72 4.51 12.96 8.62 8.38 1.72 2.16 1.55 3.71 0.99 0.74 0.17 1.55 0.92 1.91 0.53 0.79 1.69 0 0 0.82 0.75 1.07 0.62 0.1 1.21 1.16 0 0.76 S Cu u gham( F 1,488 426 1,062 297 286 289 296 1,456 Source: Cunningham (Forthcomi'ng) 101. Facing Disasters. When faced with systemic shocks such as a natural disaster or widespread unemployment, private sources such as friends and family still offer the most important source of support (43.3%). Table 13 shows that 15 percent believe that no-one will help them during a natural disaster and 33 percent believe that they will be alone in the event of wide-spread unemployment. Once again, the government, aid agencies and churches are not seen to be a big source of support in these times. Municipal government, however, is perceived as being a much better source of support than provincial and national government during natural disasters, and generally governments are seen as slightly more reliable during natural disasters than during unemployment. The Church is not seen as a major source of relief under either type of shock. Table 13: Sources of Expected Relief in Systemic Shocks (with weights) (%) _aW M_ CM77coym3 14.53 12.37 16.6 32.74 30.33 35.05 43.29 43.4 43.19 35.74 34.25 37.17 -9.45 9.49 9.42 5.69 5.39 5.99 12.49 14.45 10.61 3.92 5.64 2.27 4.59 4.09 5.07 3.2 4.16 2.29 1 .Sl 1.35 1.66 __ 0.58 0.56 -_-0.6 t8$l ~~~~~~3.37 4.24 2.54 _1.71 2.32 _ 1.13 0.1 0 - 0.19 2.21 2.53 21.91 33 3.39 I 3.29 I 3.49 j 1.71 I 1.78 I 1.65 1,488 692 796 1,488 692 796 includes community leaders, community council community, members of other communities ' includes private aid organizations, human rights groups, and NGOs3 includes Catholic, evangelical, and other * omitted include political leaders, political parties, educational institutions, international aid organizations, media, other 102. In terms of collective action as a strategy in normal times, there seems to be little experience in Argentina. Argentines are unlikely to have participated themselves in campaigns, demonstrations or community initiatives to influence a public decision-maker or to leverage public resources. They are also unlikely to have assigned this role to an intermediary-type go- between. 103. Table 14 and Table 15 (below) show that the majority of Argentines have never been involved in petitioning a public official and have rarely had contact with an influential public official. Thus, it would seem that collective action is not a strategy for linking between citizens and their representatives or governors in Argentina. Table 14: Frequency with which Members of the Community have Presented a Joint Petition to Government Officials or Political Leaders (%) GIeR[VS = ~~~~~46.6 ! ! ~~~~~~~~~11.3 Source: Prepared by Sofres-Ibope from survey data 34 Table 15: Level of Contact with Inflmuential People (%) Yes 10.0 No 87.8 Don't know 2.2 Yes 6.3 No 91.9 Don't know 1.8 Source: Prepard by Sofres-lbope from surey data 104. The reasons offered for not engaging in collective action during normal times are presented in Table 16. "Having no time" is the most cited reason, followed by lack of information about initiatives. 19 percent admit that they do not join in such initiatives due to lack of interest. Perhaps surprising is that only 2 percent of people do not engage in such actions because they feel that it would not be effective. Thus, the pace of life in Argentina seems to be the main deterrent for Argentines to engage in collective action as a strategy for influencing those in power. Table 16: Reasons for not participating in petitions (%) _1 ~~~~~~~~~~~44.9 19.1 = " ~~~~~~~~~~~18.9 m 2.4 Source: Ptpared by Sofies-lbope using survey data C. Levels and determinants of inter-personal trust 105. Levels of Interpersonal Trust. The comparative data within the World Values Survey finds Argentina to have low levels of trust, even when compared to countries of similar size in the Latin America region (see Table 3). This report confirms that trust is still low, and at a similar level to the results of the last World Values Survey. 106. Regional Levels of Interpersonal Trust. Table 17 shows that, as with participation, there are significant regional differences in terms of levels of trust. Cuyo has a significantly higher level of trust than other regions, followed by the Central region and the Southern region. It is interesting to note that the regional variations in interpersonal trust do not mirror the regional variations in the levels of participation. Thus, while the South had a higher participation rate, Cuyo has a higher trust level. The Metropolitan area has the lowest level of participation but not the lowest level of trust. 35 Table 17: Trust with Participation Rates by Regions NIi~na Won R'~on 2 R7"3033 43 ~o : Rin6 pi~.jst ____ 23.07 23.08 24.09 17.59 3333 18.09 24.00 EP~i:t0giZ 12.06 10.77 11.81 15.1i2 10.95 12.27 18.86 Source: Calculations based on survey data compiled by Sofres-lbope, Argentina. 107. Characteristics of Interpersonal Trust. Table 18 shows trust by socio-economic characteristics. It suggests that the rich are more likely to trust than the poor. Equally, the employed are more likely to trust than the unemployed and household heads are more likely to trust than other members of the family. Whereas women are more likely to participate than men, men are more likely to trust others. Table 18: Characteristics of Individuals Who Have Interpersonal Trust Have Interpersonal Do not have Trust Interpersonal Trust Demographics Male 51.97 43.54 Household Head 48.23 ** 41.03 Married 47.28 * 53.00 Years of Education 10.85 *** 10.14 Age 39.74 ns 40.12 Economiic characteristics Employed 55.26 *** 45.55 Unemployed 11.36 ns 14.21 Notes: Estimated means, taking into account the clustered and weighted nature of the data. Levels of significance for an F-test for difference in the means for Have Interpersonal Trust and Do not have Interpersonal Trust are reported. * Difference is significant at the 10% level; * difference is significant at the 5% level; *** difference is significant at the 1% level; n.s.: difference is not significant. 108. It might be expected that membership in organizations should be a function of income, while trust should be a function of wealth. The rationale for this expectation is that membership and income are both "flow" variables, while trust and wealth are "stock" variables. This is an interesting proposition, but any attempt to test it is bedeviled by the fact that income and wealth are arguably endogenous in a trust equation for precisely the same reasons that they are endogenous in a membership equation. Figures 8 and 9 show non-parametric (Fan) regressions of interpersonal trust and log per capita income (Figure 8), and interpersonal trust and a measure of wealth (Figure 9). Like the simple bi-variate correlations, the non-parametric regressions suggest that more trust is associated with higher levels of per capita income, as well as higher levels of wealth. 36 Figure 8: Probability of having interpersonal trust and log per capita income .6- .C 0) 98% .4 - 02 O2 -., E 0 3 4 5 le 7 Log of per capita income Figure 9: Probability of having interpersonal trust and wealth .45 .4 0 .4 02 .35 - -CL (- E .3- .25 - 1 2 3 4 Wealth Index 37 109. Because it is unclear whether income, wealth, or neither one of these variables should be in the trust equation, we report three sets of trust regressions in Table 19. The regressions in the first column (i) exclude income and wealth, while the regressions in the second column (ii) include measures of log income per capita and its square, and the regressions in the third column (iii) include measures of the wealth composite and the squared wealth composite. Once again, the sample is limited to those aged 16 and over, and all regressions include a vector of cluster dummies to control for unobserved cluster-level heterogeneity. Estimation is done by probit. The table reports the estimated probability derivatives at the mean with standard errors corrected for clustering.20 Table 19: Determinants of Interpersonal Trust, Various Specifications (i) (i)(ii Age -.0030(-0.84) -.0031(-.085) -.0030(-0.79) Age squared (x100) .0029(0.84) .0021(0.61) .0029(0.78) Years of education .0120(3.80)*** .0056(1.47) .0073(2.00)** Male .0174(0.59) .0035(0.11) .0163(0.51) Married -.0563(-2.18)** -.0539(-2.02)*** -.0712(-2.88)*** Household head .0781(3.06)*** .0847(3.36)*** .0837(3.09)*** House owner .0306(1.12) .0311(1.17) .0224(0.79) Log income - .2582(2.92)** Log income squared (x1OO) -1.9528(-2.25)** Wealth - .1400(2.51)*** Wealth squared (x100) - -2.520(-1.95)*** Observed probability of interpersonal trust .334 .334 .334 Predicted probability of interpersonal trust .325 .322 .324 Pseudo R-Squared .0671 .0758 .0708 Number of observations 2,088 2,088 2,075 Notes: Dependent variable is the probability of interpersonal tnust. Z-statistics corrected for heteroskedasticity and clustering are reported in parentheses. * Significant at the 10% level; ** significant at the 5% level; *** significant at the 1% level. 2( There is here no issue of larger and smaller sample sizes, as the trust questions were asked only of the respondent, and not of other household members. 38 110. The regressions suggest that more educated people are more likely to trust: an additional year of education increases the probability that a respondent will have interpersonal trust by 1.2 percent (column (i)), although the effect of education is no longer significant when the income variables are included in the regression. Household heads are also more likely to trust, while being married is associated with significantly lower levels of interpersonal trust. Controlling for other covariates, both the income and wealth measures appear to be significantly associated with trust. Trust rises with income, albeit at a decreasing rate, and peaks at about the 95th percentile of the distribution. Trust also rises with wealth, also at a decreasing rate, and also peaks at about the 95t percentile of the distribution. These results should be interpreted with a great deal of caution, however, as they could be biased because of the endogeneity of both the income and wealth variables in the trust equation. 111. Interpersonal Trust and Participation. Figure 10 suggests that there is a significant relationship between those that participate and those that trust others. Simple bivariant cross tabs show that Argentines that participate are significantly more likely to trust, or, put differently, those that trust are more likely to participate. While this finding does not show causality, it does suggest that the act of participation and the value of trust are intrinsically linked. Figure 10: Levels of Interpersonal Trust by Participants and Non-Participants _ 40% 12 300% - *Ri 25%- o 20% 0j 15% 2l 10%Z & 5% - -0% Non- Participants Partcipants Source: Calculations based on survey data compiled by Sofres-Ibope, Argentina. 112. Interpersonal Trust and Homogeneity of Membership. In addition to the correlation between general participation and trust there also appears to be a positive and significant correlation between the level of heterogeneity among members of participants and inter-personal trust.21 Figure 11 shows that those that participate in organizations with members who share similar characteristics are less likely to trust than those that participate in organizations with more diverse membership. Thus, it would seem that organizations that enable participants to meet people from "other walks of life" create higher levels of interpersonal trust than those organizations that group "like with like." 21 A homogeneity index comprising same sex, type of job, education level, income and type of problem faced was used. 39 Figure 11: Levels of Interpersonal Trust for Participants in Homogeneous and In~~~~~~~~~~ Heterogeneous Organization i..4Q% - 40%-- 30% 0 20%- *10% - 0%---- Heterogeneous Homogeneous Source: Calculations based on survey data compiled bylSofres-lbope, Argentina. Note. The differences between trust and homogeneity and heterogeneity is significant at the 5% level. 113. Interpersonal Trust and Participatory Decision-Making. Similarly, there is a significant relationship between, on the one hand, organizations that engage members in the selection of their leaders (Figure 12) and participatory furnd management and, on the other, levels of interpersonal trust.22 Thus those organizations which engage their members in choosing leaders and in managing the group's funds tend to produce a significant positive effect on the members' level of inter-personal trust. This suggests that horizontal organizations tend to promote more interpersonal trust than hierarchical ones. Figure 12: Levels of Trust of Members that Participate in the Selection of Leaders 50% ~ 40%1~~~~ _ _ _ 30% 2 30% 20% * 10%I'_ 0%1 Participate in Selection Non-Participate in of Leaders Selection of Leaders Source: Calculations based on survey data compiled by Sofres-Ibope, Argentina. Note. The differences between interpersonal trust and participatory selection of leaders is significant at the 5% Levels. 22 However, it should be noted that a positive relationship between trust and participatory fund management is not as significant - probably due to the size of the sample. 40 114. Interpersonal Trust and Solidarity. Table 19 suggests that levels of solidarity in Argentina seem significantly higher than levels of trust. However, rather than being a significant finding, this likely underlines the methodological difficulties in measuring intangible values such as solidarity. When respondents are asked whether they would be willing to contribute time to a project even if it did not benefit them, 74 percent say "yes". However, when they are asked whether they would contribute financially to a project which did not benefit them directly, they are much more reluctant (16.3%), as illustrated in Table 20. When asked a realistic hypothetical about who they would leave their children with in the case of an emergency, the level is similarly low (23.7%), as illustrated in Table 21. Thus the type of question used to measure values clearly matters and questions based on real behavior at the individual level or requiring a response which implies commitment of economic consequence seem to be more reliable. Table 20: Predisposition to Contribute to A Project from Which One Does Not Benefit Directly eT.Ypeofic'Miriu-t 'iL . 5rP,e'r& tGe4F*M_ In Yes 73.5 No 16.3 Don't know 10.2 Yes 28.5 No 50.8 Don't know 20.7 Source: Calculations based on survey data compiled by Sofres-lbope, Argentina. Table 21: Trust with Participation Rates b % of Po pulation by Region N 165801I" Rf i ~o~ R~1i~~To ~ ontr VRegion I~~~~Mtoo Zh1¶ B 4 r17WfE 6.26 4.70 6.81 7.:3 10.76 4.84 7.71 IaisAnI ; 23.07 23.08 24.09 17 59 33.33 18.09 24.010 PWirciMy 12.06 10.77 11.81 15 12 10.95 12.27 18.8( SN'thOiUr 58.61 61.45 57 29 59 46 44.96 64 80 49 43 Source: Calculations based on survey data compiled by Sofres-Ibope, Argentina. See text for definitions. D. A Baseline for Participation and Trust by Region 115. The levels of participation and trust by region make it possible to produce a basic map of social capital in Argentina. While it would be ideal to produce a composite index, illustrating the combined levels of participation and trust by region, the problems in creating such an index as a tracking tool have led us to present our findings by single components. The problerms encountered in creating a single index include combining the value of a stock with the value of a flow, assigning a value to an index with no obvious units, and standardizing the index while tracking changes of individual components over time. Thus, the presentation of the components by region in Table 21 and Figure 13 illustrate: (i) those that both participate and trust and thus have a stock of social capital accumulated and continue to invest in their social capital for the future, (ii) those that have a stock of trust but are not participating at present and are thus 41 unlikely to increase or maintain their stock at the present level, (iii) those that participate but have not, as yet, accumulated any trust - perhaps because the experience was not productive or it has yet to have an effect on their levels of trust, and (iv) those that neither participate or trust and who are therefore unlikely to change their level of social capital over time. 42 Figure 13: Trust and Participation by Region 800% 8&8% 0% 0% 'S. I ~~~~~~~~~~~ P..'~~.... ...... .... 4001 - 10%- alab~~~~~ ~ ~ ~~~~~~~ -t ~ VA'~ ~n EO '6L' 10% 40% 00% _%i .'-. . . . ... . . Metropolitan ................ .............. .....i 4 i.5 0D% -4~~~~~~~~~~~~~ O m .. .. ...... lVW JUL~~~~~~~~~~~~~~~~~~~~.... .............sxz Iw ll ~Merotiia 0% ~~~~~~~~~~~~~~~~~~~~~~~~~~41 0% r . . X b ,,, .2 *1T% 100 a I 4 \ ........~~~~~~~~~~~~~~~~~~~~~~~~~~~~~... ...... .. . w I -I w0 1 11 . I 0°%r - - =--S-- 43 - ~ ~ 40%7 24.1% F- I0%i ~ 1 - | | ba*~~~~~~~TitA IwIob§OI P.flkl.c&Sa NotWlrwn , 43 116. The above findings provide an interesting 'snapshot' of national and regional levels of social capital. Cuyo has the highest overall level of those that participate and trust, followed by the Northeast. The Metropolitan region and the Northwest have the lowest levels of those that participate and trust. These findings can serve as a baseline while the survey as a tool for tracking how these levels change over time. E. Enabling Environments and Social Capital 117. Many recent studies have suggested that there is a strong connection between social capital and its environment. An enabling environment for social capital would be composed of high levels of general trust among citizens in their institutions and leaders, transparent and effective governance and rule of law, and levels of participation and interpersonal trust at the community level. Since the final section of the survey allows to measure respondents levels of general faith in, for example, other Argentines and Argentine public institutions, this report is able to test this connection for Argentina. 118. Characteristics of Broad Trust. As noted earlier, levels of confidence in public institutions are so low that it can fairly be said that the public sector is facing a severe credibility crisis. The findings also suggest that within this general climate of mistrust of the public, the poor, women and the unemployed are significantly less likely to trust public institutions than the rich, men or the employed. The following section presents our findings on relationship between the environment and levels of social capital. 119. Broad Trust and Participation. In addition to gender, income and employment status, participation in civic organizations is a significant predictor of not only interpersonal trust but a broader trust of ones fellow citizens. Those that participate are significantly more likely to have higher levels of general trust at the macro level, as illustrated in Figure 14, 89 percent of participants answered "yes" when asked, "do you believe that Argentines are generally trustworthy people". Only 85 percent of non-participants were likely to trust their fellow citizens. Figure 14: Levels of Macro Trust for Participants and Non-Participants 100%- 5/_ 80%- 60% -- 40% - 20% - 0% Non Participants Participants Source: Calculations based on survey data compiled by Sofres-lbope, Argentina. Note: Differences in levels of general trust between those that participate and those that do not are significant at the 5% confidence level, or higher. 44 120. Broad Trust and Interpersonal Trust. Figure 15 suggests that those that have higher levels of trust in public institutions also tend to have higher levels of interpersonal trust. When asked to characterize a series of public institutions as either very effective, quite effective, or not effective at all, those respondents that considered a broad range of public institutions effective also tended to havb significantly higher levels of interpersonal trust. Figure 14 presents the levels of confidence in a broad array of public institutions for those who have interpersonal trust and those who do not. Figure 15: Levels of Interpersonal Trust and Trust in Public Institutions 950/ %r 88% ~~~91% 10 90%- 9/ 880/ 87-/o t I ;0 ~~~~9O% ~ ~ ~ ~ ~ 1 a85% X I%_ |1 7 80% =650/ = !!X Judges Deputies Municipal Judicial President - Chamber Councils System I E Do Not Trust 1D rs Source: Calculations based on survey data compiled by Sofres-lbope, Argentina. Note: Differenices in levels of trust in public institutions and interpersonal trust are significant at the I% level (*), 5% level (**) and the 10% level (*) 45 46 V. CONCLUSIONS AND RECOMMENDATIONS 121. The main aim of this report is to begin a dialogue on the ways of promoting social capital in Argentina in response to the growing perception of social exclusion. But to do that, there was the need to undertake a baseline assessment of the levels and determinants of social capital in Argentina. The idea is to use this information-in a variety of ways--to encourage the creation and strengthen social capital. The study starts with a regional and national baseline for social capital, which can be used as a tool for monitoring social capital in Argentina in the future. This report has also provided some additional thoughts for the design of surveys measuring both participation and trust at the micro and macro level. However, while the Argentina survey clearly sheds some additional light on the challenge of measuring trust, there are still methodological challenges in measuring social capital.23 The Main Conclusions and Interpretations 122. An effort has been made to generate a profile of social capital in Argentina and to interpret these findings in relation to Argentina's social, political and economic context. This analysis has generated some hypotheses about social capital in Argentina and, indeed, there may be alternative interpretations and explanations to those offered in this report which need to be assessed. In particular, it is important to continue to ask why the situation is as it is in Argentina and what can be done about it. It is also important to take into account the differences by region, gender, education, and income levels which are related to social capital, as well as the form of social organizations and their characteristics, in the design and implementation of social programs and actions. The study's main conclusions and interpretations are presented below. A. Social Capital is Low 123. Despite the growth of the NGO sector and specific examples of successful social organization, overall social capital is relatively low in Argentina - even in comparison to other developing countries. Less than 20 percent of the population participates in any form of group or organized action. Collective action is rarely used as a day-to-day strategy for engaging politicians or leveraging support. 23 In particular, this report has confirmed the challenges in measuring values such as trust and solidarity, as well as reasons for or not participating. Future surveys should consider confining questions on values to experiences at the individual level whereby responses require some commitment of resources. 47 B. Associations are often Religious Based 124. Argentines who participate regularly do so in a broad range of secular and religious organizations but the majority are religious based. Religious groups make up 47 percent of the total amount of organizations in which Argentines participate. There are a broad range of self- help groups but these are not seen as playing a major role in helping communities to manage risk C The Family Counts the Most 125. Even when faced with shocks, Argentines turn to their immediate family before turning to friends and other members of the community. This is an important factor in designing and implementing social protection schemes. D. Regional Differences are Important 126. There are significant variations in levels of social capital among the six regions - although all levels remain relatively low. At an aggregate level, Cuyo has the largest proportion of the population that both participates and trusts, followed by the Northeast and the South. The Metropolitan area and the Northwest region have the lowest aggregate levels of participation and trust. This would point to regionally differentiated strategies to promote social capital in Argentina. E. These Differences Need to be Better Understood 127. These regional differences open the door to ask why there are significant differences within the same country among the populations. Since we find that micro and macro levels of social capital are inter-related, it is likely that other community level variables also play a role in determining micro levels of social capital. One hypothesis is that the history of state-society relations and forms of governance in any one region affect the levels of micro and macro social capital. This hypothesis assumes that regions with a tradition of successful activism and solidarity are more likely to have higher levels of confidence and trust at the macro level which both promotes, and is strengthened by, participation and interpersonal trust at the micro level. 128. Since the region of Cuyo is home to C6rdoba,and Mendoza, two provinces that stand out in Argentina for their historically high levels of activism, collective action and solidarity, there may be some truth to this hypothesis.24 It is likely that the longer history and relative influence of civil groups in these cities, and their continued capacity and cautiously cooperative relations with public institutions play some role in providing-a climate conducive to the relatively high level of social capital in the Cuyo region (10.8%). In the same way, state-society relations and other 24 C6rdoba was the origin of several pivotal social movements including the University Reform Movement in 1918 which resulted in the national educational reform and the Cordobazo uprising ultimately toppled the regime headed by President Onganfa. Mendoza was settled by immigrants who, through cooperative efforts, created what is today a successful regional economy based on wine-making. Also see Noguira in Reilly (1995) for a case study on state- civic organizations relations in Rosario. 48 political and institutional determinants may help to explain why the Metropolitan region has the lowest percentage of the population that both participates and trusts (4.7%). 129. There, nevertheless, could be other explanations to these differences, not explored in this study. One of these may be the nature of the communities in the Metropolitan area: some of these are well-established with long traditions, but others are the creation of rapid urban spread, and public housing and urban development policies which have re-created communities. Another factor may be patterns of migration to the Metropolitan area by those from other provinces, or even other countries, who do not share many common bonds as communities in other parts of Argentina. 130. Also, it may be that income inequality-not just income levels-is a factor. While data are scarce, it may be that the Metropolitan area is home to wider income variations than other parts of Argentina, and the associated contrasts in life-style and socialization, work places, etc, may inhibit the formation of social capital, particularly, bridging and linking forms. 131. Or, another reason may be simply questions of size and time: Greater Buenos Aires is extremely large, and overshadows all the rest of the country in terms of concentration of population. Its size entails large amounts of time devoted to travel, to and from work. These are just some of the questions which might be asked to understand better the results. F. Links with Poverty, Gender and Education 132. The better-off, educated, employed Argentines who are also household heads are the most likely citizens to trust and participate. If we accept the hypothesis that these constitute "social capital" assets which the poor do not have, this is another example of inequality between the rich and the poor. Women are more likely to participate than men, but men who participate are more likely to be leaders which may reflect gender-based stereotyping. G. The Poor Participate to Survive while the Rich Participate to Advance 133. The poor are less likely to participate than the highest quintiles but seem more likely to participate than the middle class. This suggests that the poor turn to participation and collective action as a coping strategy, while this is not necessary for the others. The poor are significantly more likely to participate in homogeneous groups with other poor people. They also tend to belong to bonding organizations which bring them together to solve their own problems internally with the resources available to them, rather than introduce them to new resources, information, or spaces for influence. The levels of participation and the linking dimension of the social capital of the rich, on the other hand, suggest that participation is used as a strategy for the rich to maintain or increase their high levels of welfare. H. The Poor Don't have a Chance to Lead, Control or Make Decisions 134. There are important differences in the internal characteristics of the poor's organizations when compared with the organizations of the better-off. The organizations of the poor are more hierarchical and thus poor "participants" are less likely to participate in the 49 internal decision-making of their organizations, and less likely to exercise any control over the group's funds. These characteristics are an area of concern for those engaged in poverty reduction in Argentina. Their limited active participation in the management of the group is likely to mean that the experience of participation is not as rewarding, does not provide the opportunity for them to learn new skills in organizational management and is less likely to lead to the development of inter-personal trust. L Trusting Each Other and Trusting the State 135. The enabling environment and micro levels of social capital are strongly related in Argentina. Argentines who participate in their community or trust their neighbors are more likely to have confidence in the ability of the government and other public institutions. Similarly, those that have high levels of confidence at the macro level - either in other Argentines or in a broad range of public institutions-are more likely to participate in organizations and trust others. This confirms findings of earlier studies, which find that there is synergy and complementarity between micro levels of social capital and an enabling environment. It also confirms that providing an enabling environment for social capital through public action is key to the promotion and scaling up of social capital. A Broad Agenda for the Future 136. The following recommendations are laid out as the broad agenda that could be followed in Argentina to strengthen social institutions and promote social capital. Because they are based on a pioneering survey instrument with little additional information available for review on social capital in Argentina, these are offered as suggestions for consideration and debate. More work is needed to understand operational issues of how best to promote social capital at the micro level and prevent obstacles at the macro level within the overall goal of: (i) providing a climate for social capital; and (ii) promoting values and actions to enhance social capital. A. Creating an Enabling Environment for Social Capital 137. The contributions that social capital can make to economic activity, poverty reduction, risk management, and building unity during hard times are significant. Governments are important actors in providing a climate which is conducive to the development and scaling up of micro-level social capital. In Argentina, the government might consider the following steps to provide a more enabling environment for social capital: / Increase Accountability of Public Institutions. Public institutions face a serious credibility crisis in Argentina. This report also finds that low levels of trust in institutions is correlated with low levels of social capital. Thus, creating an enabling environment for social capital must include addressing the credibility crisis in public institutions. One way might be to introduce public accountability mechanisms as an approach to. service delivery in the public sector, as suggested as part of the Government's Modernization of the State program and Citizens' Charters. Such mechanisms should: (i) provide reliable information to citizens about public programs, their results and their resources, (ii) engage citizens as service users - encouraging them to provide feedback on services, and (iii) take their opinions into account 50 when planning sectoral policies, programs and budget allocations (iv) provide channels for recourse when money is misallocated. Specific tools such as user score cards, similar to those used in India, public budgeting processes such as the one carried out in Porto Alegre, Brazil and local participatory development planning such as the effort presently underway in the Southern Andean region of Peru provide specific examples of how this might be done. v Building Capacity Around Effective Decentralization. Decentralization can be a powerful strategy for ensuring the participation of the people who have the information and incentives to make decisions best suited to their needs. However, decentralization is not a fix-all, and it also offers challenges in countries where local elites continue to dominate decision making processes. Nonetheless, decentralization does provide an additional opportunity to build local capacity for public-private interaction. Decentralization, done well, also has the potential to make state institutions more responsive and attention needs to be paid to promoting values of mutual respect between local governments and other actors. Thus programs which provide local governments with the opportunity and capacity to interact with civic organizations as well as programs that build the technical skills of citizens to engage in an informed way in public processes are likely to improve the environment at the local level. Educating for Civic Action. Social capital requires concerted actions by governments and citizens alike. The history of civic action in Argentina has been characterized by an ambiguous line between public and private action, a history prone to clientalism, and politicization of civic organizations. At times, independent civic action seems to has grown, only to be subsequently crowded out by public policy or co-opted by the state. It is not therefore surprising that citizens in Argentina are skeptical or unsure about how to adjust to the independent exercise of civic responsibilities and rights. Changing deep-rooted values and practices is a process-often a long one-which begins with education and practice. Education is an important determinant of social capital in Argentina, and schools and universities are well perceived by the majority of Argentines (see Table 2). Ensuring access to quality education and using the education curricula to begin mainstreaming, a better understanding of civic rights and responsibilities as well as the complementary nature of public and private action offers an important starting point in investing in social capital in the immediate and long-term in Argentina. Schools are important centers of socialization in both what they teach, how they teach it and the space they offer for community interaction. Designing a well thought-out civic component for the school curriculum and promoting joint control of schools by parents and teachers might help to initiate values and actions of "civicness". A start has been made on this with, funding from the World Bank, via Conciencia, to train teachers in civic education. In addition, using schools for young people to interact and engage in extra curricula activities in safety, provides a space and opportunity for the creation of social capital in communities where violence may act as a deterrent. The Pro-Joven project is presently experimenting with this approach by using the school as a safe space for children and adolescents to engage in stimulating activities during their free time. The school thus serves as a safe haven in communities where drugs and violence are rife, as well as a space where activities which are complementary to and enrich the curricula can be carried out while allowing children to socialize at the same time. 51 / Taking Stock of Social Capital in the Design of Public Policies for Social Protection. Citizens generally turn to private sources-their own reserves or their families--when faced with crisis. The poor are more likely to have bonding social capital, which is generally used to protect them against shocks and pool risks. This is not surprising because public social protection programs are traditionally weak in Argentina, and citizens have learnt to cope by themselves. But there are limitations to the effectiveness of these private social coping mechanisms and room for a more public action in this area. In the event that public policies were to focus on social protection, it would be desirable that they first take stock of existing responses to ensure that new interventions complement and strengthen private ones, as well as learn from the strengths and weaknesses of these traditional mechanisms. This is part of the agenda with the Government to consolidate and strengthen social programs. B. Promofing Social Capital 138. At the micro level there is a role for both governments and civic organizations. Governments can provide infrastructure and information that promotes social capital as well as investing in the independent, organizational capacity of their citizens. Civic organizations in Argentina also have a key role to play in helping to change old values and traditions towards a new "civicness." Through their own activities and their interaction with government programs, civic organizations can help to contribute to a culture of information, trust and active citizenship. V Creating a Culture of Reliable and Free-Flowing Information. This report finds that lack of information and knowledge about existing organizations is a major constraint. It also finds that lack of information about public sector activities creates a feeling of distance between citizens and the public officials who are elected to represent them. The role of the media in promoting a culture of reliable information and informed debate is potentially key in Argentina. The media enjoys high levels of credibility among Argentine citizens, and is therefore a potentially powerful tool in the quest for more better information exchange between the government and citizens. Thus, the media could play a helpful role in disaggregating information on local and central budget processes and, in doing so, provide citizens with the necessary tools to play a more proactive role in the budgeting process with their representatives. The government and civil society organizations also have a role to play in promoting more and better quality information. Governments can help by publishing information on public programs, beneficiaries and impact on web pages. Civil society groups, in turn, can play an important role in helping disseminate information in easily accessible forms. In addition, they can provide independent research and assessments to inform dialogues with public officials. Activities such as those being carried out around the budget process by Poder Ciudadano provide examples of activities that intermediary organizations can play in promoting an informed debate among a broad number of citizens and their elected representatives. The proposed steps by the government to provide access to information-via the Cristal webpage25 -and to make information on public procurement and contracting available would also serve as major advances in building trust. Finally, there 25 http://www.cristal.gov.ar 52 may be a role for the private sector and international organizations to help connect remote populations to timely information by providing access to technology and tele-centers. V Investing in the Organizational Capacity of the Poor. The poor, in particular, have low levels of participation and trust. In addition, they participate in organizations which help them to get by, but have little potential of leveraging important impacts for them over the long-term. Organizations of the poor that cannot reach out beyond their communities to decision-makers, those who control information and sources of funding can act are limited. While these organizations can serve as important safety nets they are unlikely to become agents of transformation. Promoting the capacity of the poor's organizations to network, exchange and access information and influence public decision-makers is one way of ensuring that the organizations of the poor help them to "get ahead" as well as just "get by." External agents may help to transform the poor's bonding social capital into bridging and linking social capital by selecting carefully the type of organization and activity funded. This may be done by providing incentives to representative, grassroots organizations that have traditionally played a bonding role and encouraging them to bridge and link by networking with other organizations and entering into contact with decision makers and other influential people beyond their organization. This might be done through capacity building programs that provide the poor with information, tools and opportunities to interact with others. Supporting existing bonding organizations of poor people to make this transformation to bridging and linking organizations may also be done by intermediary type NGOs. However, such intermediary organizations must take care to invest in the capacity of the existing, grassroots organizations, to allow the organization to represent itself directly and avoid creating dependency on new forms of organizational intermediaries. While civic organizations can play a helpful intermediary role in these exercises, they must recognize that they often enjoy privileged access and skills and, if their role as development agents is to be realized, they must put these at the service of local communities. Finally, measures such as providing the poor with the necessary infrastructure - both telephones and transport - to increase their connectivity would also increase their linking potential. Urban planning measures which promote shared space through community centers, as often supported under the Bank-financed FOPAR program also have the potential to bring people together across traditional barriers. Good urban transport can also help people to mnix and break down traditional barriers. Moving Forward 139. There is clearly an urgent need for a unifying force, mobilizing all of the capacities - including civil society- to overcome the challenges Argentina is facing. Drawing upon the accumulated stock of social capital, to manage and pool risks, to seek common solutions, to create the cohesion in society, to curb self-interest and to pursue the common good, would no doubt contribute to both coping with crisis and laying the base for sustained, equitable growth, as seen in other countries. 140. Unfortunately, what we know about levels and determinants of social capital in Argentina is not encouraging, and this report suggests that Argentina lags behind other countries in the accumulation and formation of this form of capital. It is therefore unlikely that the conditions for 53 exploiting this role of social capital are in place at present in Argentina, perhaps at a time when needed the most. Particularly discouraging is the level' and nature of the social capital possessed by the poor. There are low levels of social capital at the micro level, and significant obstacles to improving this within the policy environment. 141. Moving ahead, parallel steps are needed to both promote social capital and improve the enabling environment. In terms of creating an enabling environment, government has an important role to play. And, at the micro level, both government and civil society actors have a role to play in promoting actions and values that enhance social capital. The international community can also play a role of supporting, these local actors in addressing the low levels of social capital while continuing to help track trends. 54 END-NOTE Discussions on the Report with Social Actors in Argentina This report was prepared during the calendar year 2000, based on survey data from May- June 2000. In September 2001, a draft version was disseminated among opinion leaders, public officials, civil society organizations, academics, pollsters, mayors and legislators in Argentina. This dissemination was undertaken in the hope that the report would add or promote new issues for debate at the national level on the subject of social capital and poverty reduction. The proactive approach to this exercise provided the authors with specific feedback, and as a result, some revisions were made before the publication of this report. Even more interesting was the amount and depth of debate that ensued around the dissemination meetings carried out to discuss the draft report. In particular, participants emphasized the existence of obstacles to build social capital through government-funded projects, and the challenges that civil society sector faces in providing a more enabling context for the promotion of social capital in Argentina. In an attempt to capture these additional insights, a summary of the participants' comments is provided below. The Context for Social Capital in Argentina The feedback from the dissemination meetings confirmed the report's assertions that the context for building social capital in Argentina is a difficult one. However, despite this, there was an overwhelming consensus on the need to invest in social capital in Argentina now more than ever. Much of the feedback on the environment for promoting social capital brought up the issues mentioned in the report such as: (i) authoritarism in Argentina and its impact on long-term processes such as promoting social capital; (ii) the lack of institutionality and respect for the rule of law, as evidenced by the recent house arrest of prominent officials and politicians; (iii) a heritage of state-society relations that are instrumental to provide immediate benefit, but are not part of a broader social contract and; (iv) the lack of systematic flows of timely information between citizens and their representatives, recently exemplified by the proposed 2001 budget, which was sent from the Executive to Congress with little detail, and therefore provided no basis For ensuring transparency or accountability in public actions. In particular, the report's iuggestion on the collective sense of disappointment that Argentina is not where it should be, Jespite its best efforts, was strongly reiterated in many of the discussions. Complementary comments supporting the notion that the context for promoting social capital in Argentina is a difficult one were also expressed. These included: high levels of unemployment that result in citizens no longer active in the workplace, the diversity of backgrounds among poor people, which tends to cause segregation between "new" and "old" poor rather than uniting them, sensationalism in the media that provokes uncertainty and fear, and the role of violence in deterring citizens from leaving home to engage in community activities. Finally, many assertions were made about the role of macro economic policy in providing an enabling or disabling environment for social capital. Some suggested that the market model adopted in Argentina has created a culture of individualism, with winners and 55 losers who are reluctant to participate in activities that favor the broader community. While little evidence was offered by way of justification, the role that macro economic policy plays in setting the stage for social policy and social interaction was repeatedly raised, and participants insisted that social capital is not likely to be promoted throughi social policy alone. There was a sense of urgency to begin to invest in Argentina's social capital. In particular, high level staff from the Ministry of Employment and Social Security and the Ministry of Social Development recognized that investing in social capital becomes an explicit objective in social programs. Social capital might be promoted through existing workfare, education, health and safety net programs, by incorporating specific social capital objectives, providing the staff and the financial continuity to invest in local organizational capacity, selecting organizations for promoting social capital activities addressed to the poor and carefully monitoring the impact of programs at local levels. Challenges to Promoting Social Capital through Public-financed Programs One of the most rewarding discussions was held with high-level, technical directors of programs from the Ministry of Employment and Social Security. During the meeting, a candid discussion of the present obstacles to promoting social capital through public programs took place. While some focused on the lack of political will among political appointees, other obstacles were much more operational in nature and provided some interesting food for thought. Some of the difficulties were perceived to stem from broader problems in the design of Argentine social policy such as: duplication and proliferation among programs tend to disperse populations groups, and the short-term and insecure nature of funding which acts as a disincentive to develop long-term relationships with community organizations. Other obstacles identified were more administrative in nature: the fast turnover of program staff makes it difficult for the community to engage in meaningful relationships, and accounting procedures prevent ministries from investing in intermediate or network structures. Finally, a third set of obstacles related to the design of social programs which many felt was more hierarchical and less participatory than other programs, such as those implemented in the past for adult education.26 The lack of diagnostic tools that take stock of existing organizational capacity in the design stage of programs, the specific program objectives aimed at creating local organizational capacity, the selection criteria for organizations and activities that proactively promote social capital and the indicators that measure the impact of programs in supporting the strengthening of creation of social capital were all cited as examples of obstacles that prevent a more systematic approach to promoting social capital through public-financed social programs. As a result, officials suggested the following recommendations: the incorporation of specific objectives and indicators for monitoring the impact of programs on the strengthening of social capital, the re-incorporation of participatory institutional diagnostic tools in the design stage of programs, an inter-ministerial agreement or agency to better coordinate a holistic approach to social policy, agreement on the rationale for participation, criteria for investing in local organizational capacity and incentives for project staff who are willing to accompany 26 The adult education programs that used popular education techniques during the 1970s were cited as an example of social capital-building programs in the past. 56 communities over the long-term. These recommendations were considered particularly relevant for the Ministry of Employment and Social Security which is, for example, presently engaged in promoting regional networks for the most vulnerable and unemployed. Obstacles to the Promotion of Social Capital in Civil Society in Argentina The size of the "third sector" or civil society in Argentina was pointed out as an issue of interest in several of the discussions. Recent studies, such as the one published by CEDES, estimate that there are 86,000 civic organizations active in Argentina (1995). While this number was challenged by some, it was generally agreed that the size of the sector does not reflect its ability to engage members, since the participation rate in Argentina is unusually low.27 Thus, while there is a large number of organizations, few people are participating in them. At the same time, the lack of articulation between intermediary and technical organizations with grassroots organizations was cited by some as a challenge for using civil society organizations to harness and promote social capital. Representatives of grassroots organizations felt intermediary NGOs often enjoy privileged funding and access to information, and therefore have the capacity to dialogue in an informed manner with decision makers, but they do not always represent the concerns of the poor. Some representatives felt that donors and other external agents promote the dominance of such organizations over grassroots organizations by selecting them as counterparts or funding them. The latter, which have traditionally played bonding rather than bridging roles, are unlikely to develop bridging skills and, in turn, intermediary groups can fail to provide the link to the basic needs of the poor. In terms of recommendations for the future, several suggestions emerged. These included: to encourage spaces and structures for interaction between grassroots organizations and among them and intermediary groups, to provide capacity building resources directly to said organizations for them to select their own providers of technical assistance and make an effort to engage directly with grassroots organizations, despite the "capacity gap" or "language barriers" that sophisticated policy analysts might perceive. In addition, some suggested that donors and government officials could provide incentives through funding windows and projects to ensure that intermediary organizations provide a link to local needs, and that grassroots organizations engage in activities that build their capacity to represent their own needs in influential circles. The report also clearly shows the need to better understand the impact of the World Bank's projects and activities on social capital, and some suggestions were made to establish baseline information and monitor this throughout the project cycle. In addition, others suggested that the extent of World Bank-financed reform programs can promote effective, responsive and accountable public institutions with a positive affect on social capital. Finally, two risks were identified in promoting bridging roles for grassroots organizations. The first, set within a long tradition of clientelistic ties between the Argentine state and citizens, relates to the similarities between playing a bridging role and engaging in clientelism. Then, how can governments and donors strengthen the capability of poor people's organizations to play a bridging role with decision-makers without being co-opted or used as a client by them? The 27 The numbers presented in this study have been challenged since it includes hospitals, universities etc. that are not generally considered civil society organizations that promote participation in Argentina. 57 second risk identified in promoting bridging roles relates to the balance in activities carried out by the leaders of the organizations of the poor. As one representative of an important urban grassroots movement expressed, "If I spend my time traveling and developing contacts with influential people, that is time I don't spend with my community, understanding its needs. I risk becoming an illegitimate representative of my community." The authors are extremely grateful to all participants in the discussion of the draft report for their time and candid remarks. We hope that they will agree that, as a result, the final report was improved. 58 BIBLIOGRAPHY Abraham, A. 1985. Subsistence Credit: Survival Strategies among Traditional Fishermen. Economic and Political Weekly 20(6): 247-52. Alesina, Alberto and Eliana La Ferrara. 1999. Participation in Heterogeneous Communities. NBER Working Paper 7155. National Bureau of Economic Research, Cambridge, Mass. Aronson, Dan. R. 1995. Participation in Country Economic and Sector Work. Envinronment Department Papers No. 006. World Bank, Washington, D.C. Bain, Katherine and Norman Hicks. 1998. Building Social Capital and Reaching out to Excluded Groups: the Challenge of Partnerships. CELAM meeting The Struggle Against Poverty Towards the Turn of the Millennium, April 21-22. Washington, D.C. Barnetson, Margarita.1997. H.I.J.O.S. Hijos por la Identidad y la Justicia, contra el Olvido y el Silencio. Barrio, Robert, and Jong -Wha Lee, 1993. International Comparisons of Educational Attainment, Journal of Monetary Economics, XXXII (363-394). Beard, Anne. 2000. "Social Capital and Development. " Background paper for this study. Bebbington, Anthony. 1997. Social Capital and Rural Intensification: Local Organizations and Islands of Sustainability in the Rural Andes. Geographical Journal 163(2): 189-197. Benencia, Roberto. 1992. Grupo de Andlisis y Desarollo Institucional y Social. Buenos Aires, Argentina. V, 6 Bill, Rebecca Ann. 2000. A tale of two Provinces: Countervailing Power Centers and Judicial Autonomy in San Luis and Mendoza. Presented at LASA Bombal, Inds Gonzalez and Campetella, Andrea. Historia del Sector sin Fines de Lucro en Argentina. Forthcoming Bowles, Samuel. 1999. Social Capital and Conmmunity Governance. Focus 20(3): 6-10 59 Bratton, M. 1990. Non-Governmental Organizations in Africa: Can they Influence Policy? Development and Change 21(1). Brehm, John and Wendy Rahn. 1997. Individual-Level Evidence for the Causes and Consequences of Social Capital. American Journal of Political Science, July Briggs, Xavier. 1998. Brown kids in white suburbs: housing mobility and the multiple faces of social capital. Housing Policy Debate 9(1): 177-221 Brown, L.D. and Ashman, D. 1996. Participation, social capital and intersectoral problem solving: African and Asian Cases, World Development, 24(9), 1467-1479 Burki, Shahid and Guillermo Perry. 1998. Beyond the Washington Consensus: Institutions Matter. World Bank. Calder6n, Fernando G. et al., eds. 1989. Descentralizaci6n y democracia: Gobiernos locales en America Latina. Santiago: CLACSO/SUR/CEUMT. Cambell, Tim. 1989. Urban Development in the Third World: Environmental Dilemmas and the Urban Poor. In Environment and the Poor: Development Strategies for a Common Agenda, ed. H. Jeffrey Leonard et al., 165-187. Washington , D.C.: Overseas Development Council. Carrrion, Diego et al. 1986. Ciudades en conflicto: poder local, participaci6n popular y planificaci6n en las ciudades intermedias de America Latina. Quito: Editorial El Conejo. Carroll, Thomas F. 1992. Internediary NGOs: The Supporting Link in Grassroots Development. Kumarian Press: West Hartford. Castells, Maual. 1983. The City and the Grassroots: A Cross-Cultural Theory of Urban Social Movements. Berkeley: University of California Press. Coleman, James. 1990. Foundations of Social Theory. Cambridge, Mass. Harvard University Press. Collier, Paul, and Jan Willem Gunning. 1999. Explaining African Economic Performance. Journal of Economic Literature 37: 64-111. Dahl, Robert. 1990. Social reality and "Free Markets. Dissent (spring): 224-228 De Tocqueville, Alexis. 1844 [1991]. Democracy in America. New York: New American Library. Diamond, Larry, Juan Linz, and Seymour M. Lipset. 1989. Democracy in Developing Countries: Latin America. Boulder, Colo.: Lynne Rienner Dordick, Gwendolyn. 1997. Something Left to Lose: Personal Relations and Survival among New York's Homeless. Philadelphia, PA: Temple University Press. 60 Durkheim, Emile. 1893 [1997]. The Division of Labor in Society. New York: Free Press. Easterly, William. 2000. Happy Societies: The Middle Class Consensus and Economic Development. Development Research Group. World Bank, Washington, D.C. Edwards, M. and D. Hulme. 1994. Making a Difference: NGOs and Development in a Changing World, Earthscan/SCF: London. Evans, Peter. 1992. The State as Problem and Solution: Predation, Embedded Autonomy, and Structural Change In Stephen Haggard and Robert Kaufman, ed., The Politics of Economic Adjustment: International Constraints, Distributive Conflicts, and the State. Princeton, NJ: Princeton University Press Evans, Peter. 1995. Embedded Autonomy. Princeton, NJ: Princeton University Press. Evans, Peter. 1996. Government Action, Social Capital and Development: Reviewing the Evidence on Synergy. World Development 24(6): 1119-32. Evans, Peter B. et al., eds. 1985. Bringing the State Back In. Cambridge University Press Fan, Jianquing, 1992, Design-adaptive nonparametric regression, Journal of the American Statistical Association, 87, 998-1004. Filmus, Daniel et al. 1997. El Perfil de las ONGs en la Argentina. Flacso/Banco Mundial. Foucault, Michel. 1984. Space, Knowledge and Power. In The Foucault Reader, ed. P. Rainbow, 239-256. New York: Pantheon Books Fowler, A. 1993. Non-Governmental Organizations as Agents of Democratization: An African Prospective. Journal of International Development 5(3): 325-39. Fox, J. (1996). "How does civil society thicken? The political construction of social capital in rural Mexico," World Development, 24(6), 1089-1103 Friedman, John. 1992. Empowerment: "The Politics of Alternative Development." Cambridge, Massachusetts Fukuyama, F. (1995). Trust: The Social Virtues and the Creation of Prosperity. London: Penguin. Gallup Institute. 1997. Estudio Sobre Donaciones y Trabajo Voluntario. May. Gittell, Ross, and Avis Vidai. 1998. Community Organizing Building Social Capital as a Development Strategy. Newbury Park, Calif. Sage Publication 61 Glaeser, Edward, David Liabson, and Bruce Sacerdote. 2000. The Economic Approach to Social Capital. NBER Working Paper 7728. National Bureau of Economic Research, Cambridge, Mass. Grootaert, C. 1997. Social Capital: The Missing Link? Chapter 6 in Expanding the Measure of Wealth-Indicators of Environmentally Sustainable Development, Washington, DC: World Bank. Grootaert, C. 1999. Does social capital help the poor? A synthesis of findings from the local level institutions studies in Bolivia, Burkina Faso and Indonesia. World Bank, Social Development Department, Washington D.C. Grootaert, C. 1999; Social capital, household welfare and poverty in Indonesia. Local Level Institutions Working Paper No. 6, World Bank, Washington D.C. Grootaert, C and Deepa Narayan. 2000. Local Institutions, Poverty and Household Welfare in Bolivia. Local Level Institutions Working Paper 9. World Bank, Social Development Department, Washington, D.C. Grootaert, C, Gi-Taik Oh, and Anand Swamy. 1999. Social Capital and Development Outcomes in Burkina Faso. Social Development Department Working Paper 7. World Bank, Washington, D.C. Hadenius, A. and Uggla, F. (1996). Making civil society work, promoting democratic development: what can states and donors do? World Development, 24(10), 1621-1639. Isham, Jonathan. 1999. The Forgotten Rationale for Policy Reform: The Productivity of Investment Projects. Quarterly Journal of Economics 114(1): 149-84. Isham, Jonathan. 1999. The Effect of Social Capital on Technology Adoption: Evidence from Rural Tanzania. Paper presented at the annual meeting of the American Economic Association, New York. Processed. Isham, Jonathan, Deepa Narayan and Lant Pritchett. 1995. Does Participation Improve Performance? Establishing Causality with Subjective Data. The World Bank Economic Review 9(2): 175-200. Isham, J., D. Kauffman and L. Pritchett. 1995. Governance and Returns to Investment: An Empirical Investigation. Policy Research Working Paper 1550. World Bank. Washington, D.C. Karol, Jorge L. 2000. La trama Institucional de la Ciudadanfa Urbana: Lecciones de la Privatizaci6n de servicios domiciliarios en Argentina. Presented at LASA Kaufman, Daniel, Aart Kraay and Pablo Zoido-Lobaton. 1999. "Governance Matters." Policy Research Working Paper 2196. World Bank. Washington, D.C. 62 Kesselman, Mark, and Donald Rosenthal. 1974. Local Power and Comparative Politics. Beverly Hills: Sage Kliksberg, Bernardo. 1999. Capital Social y Cultura: Claves Esenciales del Desarrollo. Revista de la CEPAL 69: 85-102. Knack, Stephen and Philip Keefer. 1995. Institutions and Economic Performance: Cross- Country Tests Using Alternative Institutional Measures. Economics and Politics 7: 207-27. Korten, D.C. 1990. Getting to the 215 Century. Kumarian Press: West Hartford. Krishna, Anirudh and Elizabeth Shrader. 1999. Social Capital Assessment Tool. Social Capital Initiative Working Paper 22. Social Development Department. World Bank, Washington, D.C. Landim, Leilah. 1988. Non-Governmental Organizations in Latin America. World Development 1 5(supplement):29-38. Lederman, Daniel, Norman Loayza and Ana Marfa Menendez. 1999. Violent Crime: Does Social Capital Matter? World Bank, Washington, D.C. Mainwarning, Scott and Timothy R. Scully. Building Democratic Institutions. Party Systems in Latin America. Stranford University Press. Moser, Caroline. 1996. Confronting Crisis: A Comparative Study of Household Responses to Poverty and Vulnerability in Four Poor Urban Communities. Environmentally and Socially Sustainable Development Studies and Monographs Series 8. World Bank, Washington, D.C. Narayan, Deepa. 1999. Bonds and Bridges: Social Capital and Poverty. Policy Research Working Paper 2167. World Bank, Washington, D.C. Narayan, Deepa. 2000. Can Anyone Hear Us? Voices of the Poor. World Bank, Washington, D.C. Narayan, Deepa and Lant Pritchett 1999 Cents and sociability: household income and social capital in rural Tanzania. Economic Development and Cultural Change vol. 47 no. 4 pp. 871-897. Narayan, Deepa, Robert Chambers, Meera Shah and Patti Petesch. 1999. Global Synthesis: Consultations with the poor. Paper presented at Global Synthesis Workshop, Poverty Group, 22-23 September, Washington, D.C. 63 North, D. (1990). Institutions, Institutional Change and Economic Performance. Cambridge and New York: Cambridge University Press. OTDonell, Guillermo. 1992. Delegative Democracy? Working Paper No. 172, Kellog Institute. Notre Dame. O'Donnell, Guillermo. 1993. On the State, Democratization and Some Conceptual Problems: A Latin American View with Glances at Some Post-Communist Countries. World Development 21(8): 1355-70. O'Donnell, Guillermo and Phillipe C. Schmitter. 1986. Transitions from Authoritarian Rule: Tentative Conclusions About Uncertain Democracies. Baltimore: Johns Hopkins University Press. Olson, M. (1982). The Rise and Decline of Nations: Economic Growth, Stagflation and Social Rigidities. New Haven, Conn.: Yale University Press. Osorio, Jorge, and Luis Weinstein. 1998. La fuerza del arco iris: Movimientos sociales, derechos humanos y nuevos paradigmas culturales. Santiago: Consejo de Educaci6n de Adultos de America Latina. Patio Mayer, Hernan. "Crisis Siste'mica de la Seguridad Ciudadana" en Archivos del Presente, Year 4, No.16, Buenos Aires 1999 Portes, Alejandro, ed. 1995. The Economic Sociology of Immigration: Essays on Networks, Ethnicity, and Entrepreneurship. New York: Russell Sage Foundation. Portes, Alejandro. 1998. Social Capital: Its Origins and Applications in Contemporary Sociology. Annual Review of Sociology 24: 1-24. Portes, Alejandro and Patricia Landolt. 1996. The Downside of Social Capital. The American Prospect 26: 18-21, 94. Przeworski, Adam. 1991. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. Cambridge: Cambridge University Press Putnam, Robert. 1993. Making Democracy Work: Civic Traditions in Modem Italy. Princeton, NJ: Princeton University Press. Reilly, Charles. A. 1995. New Paths to Democratic Development in Latin America. Boulder, Colorado. Rodrik, Dani. 1999. Institutions for High-Quality Growth: What They Are and How to Acquire Them. NBER Working Paper 7540. National Bureau of Economic Research, Cambridge, Mass. 64 Rubio, Mauricio. 1997. Perverse Social Capital: Some Evidence from Colombia. Journal of Economic Issues 31(3): 805-16. Sampson, Robert, Jeffrey Morenhoff and Felton Earls. 1999. Beyond social capital: spatial dynamics of collective efficacy for children. American Sociological Review 64(5): 633-60 Seligman, Adam. 1997. The Problem of Trust. Princeton, NJ: Princeton University Press. Serageldin, Ismail and Christiaan Grooteart. 2000. Defining Social Capital: An Integrating View. In Partha Dasgupta and Ismail Serageldin, eds., Social Capital: A Multifaceted Perspective. Washington, D.C.: World Bank Schafft, Kai and David Brown. 2000. Social capital and grassroots development: the case of Roma self-governance in Hungary. Social Problems 47(2): 201-219 Skocpol, Theda. 1996. Unraveling from Above. The American Prospect 25: 20-25 Solow. 2000 in Dasgupta, P and Serageldin, I (eds) Social Capital: A multifaceted Perspective. Washington D.C. World Bank Tendler, Judith and Sara Freedheim. 1994. Trust in a rent-seeking world: health and government transformed in Northeast Brazil. World Development 22(12): 1771-91. Tendler, Judith. 1997. Good Government in the Tropics. Baltimore: Johns Hopkins University Press. Thompson, Andres A. 1997. Non-governmental organizations and philanthropy in Latin American: An overview. International Journal of Voluntary and Nonprofit Organization V.8, No.4 pp.337-350. Thompson, Andres A 1997. Non-governmental organizations and philanthropy: the case of Argentina. V.8, International Journal of Voluntary and Nonprofit Organization No.4 pp.401-471 Thompson, Andres A. 1995. Publico y Privado: Las Organizaciones sin fines de lucro en la Argentina. Buenos Aires, Argentina. Tilly, Charles. 1981. Class Conflict and Collective Action. Ed. by Louise Tilly and Charles Tilly. Sage Publications, Beverly Hills: SAGE. Tokman, Victor E. and Guillermo O'Donell. 1998. Poverty and Inequality in Latin America: Issues and New Challenges. South Bend, IN: University of Notre Dame Press. United States Embassy. Unpublished opinion survey condlucted by the United States Embassy in Buenos Aires, Argentina. 65 United Nations Development Program, Interamerican Development Bank. 2000. Index of the Development of Civil Society Organizations in Argentina, Edilab, Buenos Aires. Uphoff, Norman. 1984. Local Organizations: Intermediaries in Rural Development. Ithaca: Cornell University Press. Uphoff, Norman. 2000. Understanding social capital: learning from the analysis and experience of participation, in Partha Dasgupta and Ismail Serageldin (eds.) Social Capital: A Multifaceted Perspective Washington, DC: The World Bank Varshney, Ashutosh. 1999. Democracy and Poverty. Background paper for World Development Report 200012001. World Bank, Washington, D.C. Woolcock, Michael. 1998. Social Capital and Economic Development: Toward a Theoretical Synthesis and Policy Framework. Theory and Society 27(2): 151-208. Woolcock, Michael and Narayan, Deepa, 2000. Social Capital: Implication for Development Theory Research and Policy. The World Bank Research Observer, vol. 15 No.2 pp.225- 249 World Bank. 2000. A Toolkit for Measuring Social Capital Poverty Reduction and Economic Management Division. Washington, D.C. World Bank. 2000. Poor People in a Rich Country: A Poverty Report for Argentina Poverty Reduction and Economic Management Division. Washington, D.C. World Bank. 2000. World Development Report 200012001. Washington , D.C. Zuleta, E. and S. Cesilini 1999 Evaluacion Social de la Autopercepci6n de los Pobres en la Argentina Poverty Reduction and Economic Management. Washington DC. World Bank. 66 Report No.: 24281 AR Type: SR