Economic growth, poverty reduction, and the role of social policies: the evolution of the World Bank's social development approach.
of social has changed. The article's main objective is to explain this change in the Bank, allowing us to answer questions about the conditions for policy change in international organizations. Is the World Bank merely adapting to the outside world, or is it engaging in innovative behavior? The case study reveals an alternative explanation for policy change in IOs other than either external pressure or internal advocacy. It shows that external and internal triggers are related and reinforce each other. KEYWORDS: World Bank, development strategy, internal advocates, international organizations, discursive change.
The World Bank has undergone tremendous change during the past decade. (1) One crucial innovation is its initiative to raise social development to the level of core issue within its development strategy, as reflected in its 2005 social development strategy. (2) While there had never been a Bankwide approach to social development before, the Bank has always been concerned with social issues. However, the meaning and operational significance of the social has changed over time. Once at the core of the development strategy in the late 1970s, social policies seemed to be forgotten over the course of the 1980s. In the early 1990s, social policies were transformed into facilitators of economic growth in order to reduce poverty, before finally returning to become the core of development today. Social policies have always been closely tied to the relationship between economic growth and poverty reduction and thus to the Bank's development strategy. The role of social (development) policies within the economic growth and poverty-reduction nexus has shifted according to the World Bank's understanding of development at different times. I argue that the Bank's discourse on the social dimension of development can be delineated into three periods. Within each period, the Bank's understanding of social (development) policy is reflected in different problem definitions and underlying assumptions and, hence, different policy responses. Thus, this article intends to describe the discursive evolution of social (development) policies and their operational consequences in the Bank from the 1980s to the current social development approach. The main objective is to understand this change in the Bank, which will allow us to draw some plausible conclusions about the conditions for policy change in international organizations (IOs).
IOs are perceived here as quasi-autonomous actors that do not merely adjust to their outside world. Contrary to conventional international relations (IR) theory, this article focuses on IO agency. While neorealism and neoliberalism explain IO behavior through the influence of external actors, their interests, and the distribution of power among them, constructivists stress the importance of norms, ideas, and beliefs that determine how IOs change their policies. Constructivism also ascribes more autonomy to IOs based on the legitimacy of rational-legal autonomy as well as on their control of information and expertise. (3) Yet constructivism places greater emphasis on the impact of IOs--their power in defining development and shaping the development discourse--rather than on processes of change within an organization. Recently, however, two alternative approaches entered the realm of IR: principal-agent (PA) models, and studies inspired by organizational sociology. Both assume IO autonomy and, thus, agency; and both attempt to open the "black box" of IOs. However, while the first places more explanatory emphasis on external pressure, in particular by member countries, the latter focuses on internal processes such as self-reflection and organizational learning in understanding policy change. I argue for a third way--one that synthesizes both approaches.
The remainder of the article proceeds as follows: The first section briefly discusses both approaches available to study policy change in IOs. The second section provides the discursive evolution of the social development approach within the World Bank, which will illuminate how and when the Bank's meaning of social has changed over time. Focusing on the ruptures in the discourse described before, the subsequent section contextualizes these empirical findings to explain why the Bank changed the way it did. This analysis will therefore allow for answers on two levels: First, it will provide policy-relevant information about the World Bank's current development strategy. Although this is only a single case study, it will, second, be possible to draw some general statements about the conditions under which external stimuli or internal processes determine policy change in IOs. According to Alexander George and Andrew Bennett, single case studies can contribute to theory building in various ways. (4) Through process tracing and multiple observation, this case study reveals an alternative explanation for policy change in IOs other than either external or internal pressure. The case study aims to show that external and internal triggers are related and reinforce each other. Change is more likely to happen in IOs that provide for specific internal conditions, such as a certain staff composition and financial autonomy.
Approaching Policy Change in International Organizations: Internal or External Triggers?
Compared to the Bank's understanding of social issues two decades ago, the adoption of its social development approach in 2005 was a significant policy change. Why? Catherine Weaver, in this issue, comprehensively describes the two approaches currently at hand in IR--the PA model and approaches inspired by organizational sociology--in an effort to explain the changing behavior of IOs. Without repeating the main underlying assumptions of PA models, what is important in this article is that PA models generally hypothesize external pressure as the driving force for policy change in IOs, while organizational sociology hypothesizes that organizations can effect policy change as a result of internal processes rather than external pressure. There are two main problems with PA models: First, they focus only on member states as the organization's principals to determine internal IO activity. Therefore, they neglect other decisive actors in the organization's complex political environment--in particular, the pressure of nongovernmental organizations (NGOs). (5) Second, in an attempt to find causes for performance gaps between the principals' will and the agent's actions, they assume that organizational interests diverge from member states' preferences. However, PA models fall short in explaining these interests. PA models assume IO interests based on their actions. Hence, PA models is no room for the possibility that large organizations such as the World Bank have their own preferences independent from the principals' will and are therefore able to initiate change themselves, or to act rather than merely react. Any step forward by the organization to commence critical self-reflection on current policies and operations is assumed to be immediately circumvented by entrenched interests within the organization. Organizational sociology, in contrast, acknowledges that external pressure from various actors in the organization's environment might be important in triggering change in IOs and that the organization's culture and interests will function as a filter through which external stimuli enter the organization and are then refracted and processed internally. It would further allow for the organization to change without external stimuli. Organizations have their own "social life" consisting of individual actors who, given specific circumstances and conditions, might engage in changing the organization's policies.
The present analysis of the evolution of the Bank's social development approach provides evidence for both explanations. The organization's material and normative environment significantly affects its behavior and policy change. This can be powerfully mediated and partly controlled by the member states. For instance, a change of key actors within the organization, such as the replacement of President Robert McNamara (1968-1981) with Alden Clausen in 1981, reflected the changing climate in the 1980s. Through the PA model, this demonstrates state preferences. However, this is not the only influence that IOs face. Other actors and events in the organization's environment are important determinants of IO behavior. In relation to social policies, NGOs and other aid agencies influenced the overall thinking of the development community. Independent external events such as financial crises or a client country's own political and socioeconomic situation also affect the notion of development. Nevertheless, the analysis suggests that IOs are not marionettes but, for different reasons, decisively shape the material and normative context in which they are embedded. For one, their interests in expanding or at least maintaining their power position can lead them to actively search for new roles and policy areas. Thus, the Bank's disposition not only functions as a filter through which external pressures are refracted, but it is also an independent source of action. Second, the Bank's own research units, policy evaluations, and operational experience provide knowledge and evidence whether policies are successful or not. Over time, they detect new issues and policy problems and thus redefine development policies. However, it is not enough that certain knowledge exists; it needs to be brought into the dominant discourse. The mixture of staff facilitates such processes.
As this brief outline reveals, policy change in the World Bank is a highly complicated process and takes place within a complex institutional and political context. To determine whether internal or external conditions have greater influence on IO behavior, it is important to trace policy change in IOs at the organizational level. Based on these considerations, the next section analyzes the evolution of the World Bank's understanding of the social from the early 1980s onward as a case study to ascertain crucial triggers of policy change in IOs. Thus, the article focuses mainly on Bank-internal discourses and processes of change while taking the organization's complex environment into account.
The Meaning of Social in the World Bank's Understanding of Development
The Bank's internal definition of social is very broad. In 2005, the World Bank adopted its first Bankwide Sector Strategy Paper (SSP) in social development. Yet the initial draft of the SSP in 2004 omits a precise definition of social development. It describes social development as "positive social change" that refers to a "transformation that better equips society to reduce poverty." (6) Social development requires a special knowledge that refers foremost to local contexts: "It requires an understanding of power dynamics, culture, and value systems, as well as the informal and formal structures of societies." (7) The final version of the SSP in social development maintains this broad definition: "Social development means transforming institutions to empower people," and institutions are broadly defined as "the set of formal and informal rules, norms and values that operate within societies." (8)
These are vague definitions. A glance at the work of the World Bank's Social Development Network (SDD), established in 1997, provides a better picture of how the Bank operationalizes social development. The SDD's activities address four core topics: social analysis; participation; diversity, vulnerability, and inclusion; and institutions. (9) Its definition of social development within the same report is, however, similarly imprecise: "The term social development has two meanings. It can refer to: improvement in the welfare and quality of life of individuals; or changes in societies--in their norms and institutions--that make development more equitable and inclusive for all members of a society." (10)
These "two meanings," so the report goes, distinguish social development from social policies. This seems to correspond with an operational divide within the Bank in the SDD and the Social Protection and Labour Market sector, established in 1996. The latter aims to improve or protect human capital and includes the protection of disadvantaged groups, labor market interventions, pensions, and other types of social insurances or income support in the form of social safety nets and social funds. These refer to "traditional" social policies. In contrast, social development embraces much broader, more cross-cutting topics, such as participation, institutions, and social capital. Thus, the understanding of social issues in the Bank refers to both of these meanings. This becomes obvious in an Operations Evaluation Department (OED) evaluation of the Bank's social development activities that operationalizes social development according to eighteen themes, including both social development and social protection. (11)
These vague and sometimes overlapping definitions, as well as operational distinctions, reflect the development of social issues in the Bank from specific social policies in the 1970s, late 1980s, and early 1990s toward broader conceptual themes at the program, rather than project, level introduced at the end of the 1990s. The content analysis of policy documents and strategy papers issued by the World Bank from 1980 to the end of 2005 provides evidence of a cumulative process. There are three periods that build on each other and are distinguishable by diverse meanings with different assumptions about social issues in development. They thus reflect distinct development strategies. First, in the early 1980s, social issues entered the discourse along with the social costs of structural adjustment. Next, through structural adjustment and its social dimension, new areas and activities came to the fore, such as social funds and social safety nets. These policies were linked with traditional social policies, such as education and health services. By the late 1980s and early 1990s, these social policies were considered facilitators of economic growth. Finally, by the end of the 1990s, the Bank acknowledged that economic growth through increasing income is not the only poverty-reduction strategy. Instead, development should include people and social norms, values, and institutions; a holistic approach to development thus emerged. In other words, these observations make obvious that social (development) policies play a prominent role within the perceived relationship between economic growth and poverty reduction. According to the specific framing of this relationship, social (development) policies were considered successively as a by-product or victim of economic development, which was assumed to automatically lead to poverty reduction (during the 1980s); as an essential means to economic development in order to better achieve poverty reduction (late 1980s and early 1990s); and, finally, as something that is defined as the core of development itself (present). (12)
The remainder of this section outlines these three distinct discursive periods. Since social development seems to play a crucial role in combination with economic growth and poverty reduction--the two main objectives of the Bank's development strategy--a better understanding of its evolution sheds light on the Bank's current understanding of development. However, the analysis does not remain on the discursive level but, in the subsequent section, contextualizes critical discourse moments in an attempt to determine what has driven the transformation of social (development) policies over the past twenty years.
The Social Costs of Adjustment
The Bank introduced structural adjustment lending (SAL) in the early 1980s. At the same time, issues of social development entered the debate, although they were initially not interconnected. In 1980, the third World Development Report (WDR) introduced economic growth and policies of human development--education, health, nutrition, and fertility--as the two main pillars necessary for successful development. Arguments about the relationship between growth and poverty reduction in 1980 were reiterated more than a decade later: "Growth is vital for poverty reduction but it is not enough." (13) The proposition that economic growth is the solution to poverty reduction is not doubted "but it needs to be carefully qualified." (14) The relationship between the two was perceived as follows: "While there is now increasing recognition that growth does not obviate the need for human development and other steps to reduce poverty, it must be stressed that the converse is true as well--direct steps to reduce poverty do not obviate the need for growth." (15) However, soon thereafter, in the early 1980s, the discourse took a different route, establishing a causal relationship between structural adjustment, economic growth, and poverty alleviation (in that order). Internal Bank discourse regarding social issues shifted to the social costs of structural adjustment. In this context, economic growth became the priority objective and social issues were subordinated, since social development was perceived to be a consequence of economic development.
The initial situation was clear, as were the solutions. Adjustment was perceived as a necessary precondition for poverty reduction based on the conviction that structural adjustment alleviates poverty in two ways: First, the reduction of market distortions would lead to a situation where "the incomes of the wealthy are likely to fall and those of the poor to rise," (16) where adjustment enhances efficiency and equity. Second, "if structural adjustment ultimately promotes economic growth, the alleviation of poverty will undoubtedly be an easier proposition." (17) Adjustment is conducive to growth, and since growth was considered to be "the most effective way to help the poor," (18) adjustment is therefore necessary to reduce poverty. Yet, the social implications, or social costs, of such programs were not ignored. Unlike the International Monetary Fund (IMF), the Bank comprehensively systematized the social costs of adjustment and conceptualized policies to mitigate those adverse effects through temporary compensatory measures that would precede the social safety nets and social funds of the 1990s. (19)
The Economic Value of Social Policies
By the end of the 1980s and in the early 1990s, it became obvious that an increase in growth rates and a decrease in poverty had not been achieved through SAL. On the contrary, in Latin America, for instance, poverty increased 27-32 percent during the period 1980-1989. (20) In addition, the distribution of income also deteriorated. In this situation, poverty entered the debate (again) and became prevalent in the early 1990s. The WDR 1990, for instance, dealt explicitly with poverty: "Poverty alleviation is what economic development is all about, and practically everything that the Bank does, either in its lending or in its policy work, bears directly or indirectly on poverty reduction." (21) The World Bank president at the time, Barber Conable, characterized poverty reduction as the "integrating theme for the many facets of the Bank's work, and it is the raison d'etre for [its] operational emphases," (22) such that "attacking poverty is not primarily a task for narrowly focused antipoverty projects, vital though these may be. It is a task for economic policy at large." (23) But what did poverty mean?
Poverty reduction had a different meaning in 1990 than it has today. Reducing poverty meant increasing the income-earning assets of the poor, as well as providing opportunities for the poor to take part in economic processes. The argumentation was that "even though the key to poverty alleviation lies in efficient economy-wide policies that promote stability and growth, there is typically considerable scope for targeted interventions to assist the poor directly.... Primary education and literacy, for example, are extremely important from both social and economic points of view." (24)
The two-pronged strategy to attack poverty outlined in WDR 1990 summarizes this approach. The first pillar aimed at the expansion of employment and income-earning opportunities for the poor and was thus concerned with the nature and rate of economic growth. Economic growth included improving macroeconomic and sectoral policy frameworks as well as developing the private sector. The second pillar sought to enhance the ability of the poor to use these opportunities by directly improving their welfare through access to health care, education, and other social infrastructure. This strategy was supposed to be complemented by social safety nets to protect the poorest and most vulnerable. It was designed "to ensure that the poor both gain from and contribute to growth." (25)
The main function of any kind of social policy in the early 1990s is therefore seen primarily in the economic value it has for societies, or, in other words, as a means to the end of economic growth. The title of a working paper published by the World Bank makes this obvious: Social Development Is Economic Development. (26) The paper argues that social development is "an excellent investment--in terms of its contribution to economic growth." (27) Social development is perceived as "good economics," and "social programs are superb investments in future economic growth" (28) and have an economic return. Social policies or social services at that time mainly meant education and health care. This is no surprise, since not only are these policies the easiest to justify in economic terms, but they are also easy to operationalize and implement using the Bank's technocratic approach to development. Immediate (quantitative) indicators are at hand, such as the percentage of primary school enrollments or the number of doctors per capita, that can serve as targets when formulating objectives and when monitoring.
The Integration of Social Development into Economic Development
Over the course of the 1990s, the approach to social (development) policies changed significantly. From the mid-1990s until the turn of the century, the Bank tried to develop a social development approach. By the beginning of 2004, a first draft of a Bankwide social development strategy paper was disseminated that tried to bundle the Bank's social activities into one framework. Many issues that appear at first glance to be new, such as participation, vulnerability, or empowerment, had been known in the organization for years but either under different labels or in specific single-policy areas. What was new was their combination within an all-encompassing approach and their enhanced operational significance as a major component of the Bank's development strategy.
The WDR 2000/2001 on poverty is a good example of this old/new combination and highlights the ascension of the social to becoming an overall theme in the development discourse. (29) The report begins with a different definition of poverty and, thus, of the problem:
The report accepts the now established view of poverty as encompassing not only low income and consumption but also low achievement in education, health, nutrition, and other areas of human development. And based on what people say poverty means to them, it expands this definition to include powerlessness and voicelessness, and vulnerability and fear. (30)
The notion of poverty as a multidimensional problem stems from Voices of the Poor, a background study conducted in preparation for WDR 2000/2001. Taking this new perspective of the problem as a backdrop, the report develops a three-pronged strategy to tackle poverty that focuses on the poor's opportunities, empowerment, and security. However, it is a combination of previously applied means and policies.
The new three-pronged strategy has two characteristics. First, the social dimension permeates all development. There is no strict distinction anymore between economic policies and social policies; both are depicted as intertwined. It is a much broader approach to development that recognizes the complexity of the problem. This links it to the second observation: that strategy is a potpourri of everything the Bank has done so far in the area of social policy and social development, brought together in one strategy and linked through three broad concepts of opportunity, empowerment, and security. The policy suggestions from the three-pronged strategy are not significantly different from those from the two-pronged strategy ten years prior. The approach previously included macroeconomic and regulatory policies, investment in human capital, and social safety nets for "the poorest of the poor." The difference lies in the depiction of the situation and in the concepts employed today. Thus, it is a change in terms of the links established among the different means to reduce poverty. Whereas the 1980s focused on economic growth leading to poverty reduction, the 1990s saw economic and social development as necessary to reduce poverty. Now it seems that social development policies and poverty alleviation are assumed to stimulate economic growth, thus broadening the Bank's understanding of development. The main difference from the previous understandings of development is that social policies and investment in human capital is not enough: "Social development is not simply a matter of social service provision. It also depends on a range of political, economic, institutional, and cultural factors, that together play a critical role in poverty reduction and social inclusion." (31) Indeed it looks like a "new paradigm of development, premised on the argument that without attention to the social underpinnings of development, it is difficult for economic growth and development to succeed--and virtually impossible for it to be sustained." (32)
External Pressure and Internal Advocacy
How can we explain these shifts in the Bank's perception of the social? Developments in the early 1980s seemed to be externally driven. The political climate at that time shifted dramatically with the election of governments of leading economic countries that favored neoliberal economic policies--for example, the United States, the United Kingdom, and Germany. These changes had consequences for the Bank. The new Bank president, Alden Clausen, gave the Bank a neoliberal drive upon entering office in 1981. These developments coincided with the economic deterioration of Latin America. Because of economic instability during the 1970s and the Bank's consequent concerns with repayment prospects, the Bank sought new solutions to deal with countries' economic problems. Once brought into the debate, macroeconomic adjustment increasingly displaced the focus on social or human development from the Bank's agenda. One quote depicts the dominant perspective in the Bank at that time. Even before the Mexican debt crisis emerged, Ernest Stern, by then operations chief, made the following statement: "In times of economic crisis, as now, the automatic response of national planners is to reduce allocations to human development (including population) in favor of shorter term investments." (33) When the Latin American debt crisis escalated, social issues disappeared from the Bank's agenda. (34) After a decade of attempts to combine economic growth with poverty reduction, attention turned to short-term objectives of economic stabilization and policy reform rather than long-term development fostered by McNamara.
The growing erosion of McNamara's poverty focus was furthered by two internal occurrences: First, the Berg Report, published in 1981, provided evidence that severe problems on the African continent were, in addition to adverse external factors, mainly due to flawed domestic policies and corruption. (35) Thus, the Bank was obviously on the right track with SAL and privatization. Second, there was growing poverty project fatigue among staff. Most often these programs had been costly, time-consuming, and unsatisfactory in their results. Pushed by the Bank's culture of the pressure to lend, many staff became increasingly frustrated, and scepticism about poverty-focused projects turned into cynicism. The poverty focus was perceived as the main obstacle to lending performance (or volume lent). (36)
In sum, in the early 1980s, conditions outside as well as inside the organization were conducive to neoliberal policies that mainly neglected the social dimension of development. Within the Bank's strategic and operational documents, references to social issues were scarce, and poverty no longer appeared to be a major problem. (37) Nevertheless, the category of the social had always been in Bank discourse, but now it connoted the social costs of adjustment. How was it possible that the Bank established the social costs of adjustment immediately after the introduction of SAL? There are two explanations. First, the Bank's mandate includes development, (38) and thus the organization has always been concerned with poverty alleviation. Prior to structural adjustment, the Bank addressed long-term poverty through poverty-oriented lending in projects concerning productive employment; improvement of physical and human assets (health, education, nutrition); access to basic social needs; reform of government agencies; agriculture projects; and urban development. It amounted to a substantial part of Bank lending other than adjustment lending. (39) Thus, expertise on social issues existed in the organization.
Second, these activities were reinforced by President McNamara, who turned the Bank into a development agency. McNamara significantly extended the Bank's scope of intervention into new areas, including social development. He searched for an approach to combine economic growth and poverty reduction, as reflected in the WDR 1980. He not only increased the lending volume of the Bank in social sectors but also restructured the organization. (40) The restructure enhanced the organizational power of the regional departments as opposed to the technical or project departments and therefore provided new possibilities for social and distributional lending. Furthermore, he established several new units working on social issues, including the Rural Development Department (1972) and the Health Department (1979). (41) Finally, in 1977, he hired a first cohort of social scientists--in particular, sociologists and anthropologists--to create a special "pioneer division" headed by Leif Christofferson. (42) In the years to come, they became very important in fostering the Bank's social agenda. (43) McNamara thus laid out the foundation for the Bank's engagement with social policies throughout the 1980s, which was possible only because the Bank is partly autonomous from its members with regard to capital investment. This is a significant difference from its sister organization, the IMF, and makes the Bank more autonomous regarding the content of its policies. Money gained on the capital market can be used to hire new staff, restructure the organization, and initiate research independently from the will of member states. It can thus be used to deliberately trigger internal pressure for change. When social issues fell out of favor in the early 1980s, a group of internal advocates tried to stop this process.
While the social focus disappeared from the Bank's discourse in the early 1980s, it did not happen without internal resistance by staff hired in the 1970s by McNamara to implement his poverty agenda. One example is Mahbub ul Haq, director of policy planning and program review since 1972, who was known as "McNamara's unofficial poverty activist in the Bank." (44) It was at his prompting, in 1981, that a task force under Vice-President Hollis Chenery (another of McNamara's recruits) was established to take stock of the Bank's activities related to poverty alleviation. The resulting report emphasized the importance of economic growth and downplayed the assumption that poverty alleviation comes with a negative tradeoff for growth. (45) It reaffirmed the Bank's poverty mission as well as made proposals on how to strengthen it. Despite powerful opposition, most notably from Ernest Stern, then senior vice-president in charge of operations, the report made it to a board seminar where it received support from many directors. (46) However, even though it was published and widely disseminated, its publication coincided with Mexico's debt default and thus did not halt the erosion of the Bank's social agenda.
It therefore seems that the Bank's policy path was mainly influenced by external forces. Economic crises and political change seemed to trigger policy change in the Bank. The Bank is a highly political entity, reflecting the political convictions of its main donor countries, many of whom elected conservative governments in the 1980s. This political climate was in line with the Bank's own research as well as its culture of loan approval. But what about the existing departmental units and staff in social areas? Both remained from the McNamara period and became increasingly embedded once introduced within the organization. All the divisions established under McNamara, such as the Health Department and the Rural Development Department, continued their work but without the previous high level of attention. Even the lending volume for the social sectors did not drop significantly. (47) There was internal resistance by a few "unorthodox" staff against the developments in the early 1980s but without success. Yet these staff members continued campaigning for incorporating social issues into development throughout the 1980s. Even though the discourse on social issues was muted over the course of the 1980s, it was then that internal advocates established the first operational policies and manuals with social content. (48) These policies and procedures became increasingly important and powerful as external critique mounted at the end of the 1980s, in particular by other aid agencies and NGOs.
The criticism centered on two different situations. First, droughts caused severe famine in sub-Saharan Africa (SSA) in the early 1980s. Public awareness was enhanced when pictures of starving children in Africa were published globally. Attention not only turned to the African crisis, but also questioned the legitimacy of Bank activities in developing countries generally. This was the beginning of massive criticism of the Bank's SAL. The most resounding voice of this critique came from UNICEF, which published a report called Adjustment with a Human Face, which argued for social policies to be incorporated into adjustment programs. (49)
The second critique came from NGOs, who increased their involvement in opposing the negative social and environmental implications of large Bank projects. The most cited example is the Sardar Sarovar (Narmada) Dam in India. This case is widely known as the project that stimulated a comprehensive process of Bank learning regarding its resettlement policies. (50) At first, NGO activities were directed at specific projects concerned with issues such as indigenous peoples and the environment. These efforts had three main consequences for social issues: First, they increased public awareness--and thus intrusive public scrutiny--of Bank engagement generally and later more specifically. Second, the environmental agenda expanded beyond concerns for the physical biosphere and quickly became a synonym for social issues. Not only would huge Bank projects have environmental impact, but they would also have social impact in terms of resettlement. Moreover, environmental problems of projects have a direct effect on people and hence are related to social issues. And third, the issue of resettlement introduced another, now famous, topic of social development: participation. In Bank project implementation at that time, the project's social impact on actual people was not considered at all, despite its enormity. Thus, carrying out a social impact analysis became part of the Bank's social agenda. In the end, this led to the framing of development as a participatory and people-centered approach and is one of the main aspects of the Bank's newly defined social development approach.
By the end of the 1980s, this loud and powerful criticism against the Bank enhanced the position of existing social staff inside the organization. These social staff had been active throughout the 1980s, employing various strategies in order to incorporate the social dimension into the Bank's development perspective. Hence, it was not NGO pressure and other critical voices alone that shifted the Bank's attention to social issues in the early 1990s. Rather, their critique found a receptive audience inside the organization. Internal advocates played a crucial role in bringing the social aspects of development to the fore. As Jonathan Fox discussed in relation to the case of the Narmada Dam, internal advocates strategically used external critique by groups such as the Environmental Defense Fund or Oxfam to foster the social agenda within the Bank. (51) This was decisively different from the IMF case, where social knowledge in the form of social staff did not exist and thus NGO critique did not resonate. In the beginning, these advocates, called noneconomist social scientists (or NESSies (52)), were McNamara's recruits.
The NESSies developed their own structures inside the Bank. (53) Very soon after their recruitment, a so-called Sociology Group was formed, led by Michael Cernea. This group met periodically to discuss specific social issues of development and invited guest speakers from outside the Bank, including NGO representatives. It usually drew twenty to fifty Bank people to brown bag lunches or presentations. (54) As Cernea says, "Bringing social knowledge into the Bank was my challenge from the very first day I joined the World Bank, my 'ToR.'" (55) As he also noted, the very language for that undertaking did not exist in the Bank at the time. But with these few social scientists, such knowledge and language entered the Bank and found materialization not only in such informal meetings as the Sociology Group but also in the publication of books, reports, and articles on the social dimension of development. (56) Moreover, these unorthodox staff members actively pursued policy change, employing a variety of strategies to promote new topics within the organization: networking with internal and external actors and building powerful allies to exert their agendas; lobbying within the organization at the board and management levels; using external pressure to (re)frame policy issues; and extending the organization's goal by reinterpretation. (57)
Thus, the driving forces of policy change in IOs were the result of internal advocacy combined with external critique by NGOs rather than member states. Both shifted the Bank's focus on the social costs of adjustment toward including social issues in the discourse, albeit with a low profile and commensurate with their usefulness for economic growth. This came as a consequence of a particular strategy of those internal reformers. Translating social issues into the Bank's economic language was the advocates' main mission at the beginning: "Yet 'selling culture' was then a counter-cultural position within the dominant economic ideology of the Bank." (58) In other words, these advocates used the benefits of social aspects as an argument for enhancing project effectiveness and later as an instrument for reaching economic growth. (59) Their next step was a strategy called "getting from project to policy level"--that is, transforming specific project practices into overall guidelines or directives that apply to all Bank activities. (60) In other words, the respective social knowledge had been present in the Bank since the late 1970s. These advocates wrote reports and led evaluations about the role of social knowledge and social policies in development. (61) However, it required a situation of decreasing Bank reputation and increasing uncertainty about the appropriate development strategy reflected in external critique to make these internal voices more powerful and to facilitate the implementation of this knowledge.
In fact, both external pressure and internal advocacy reinforced each other: The mere existence of internal advocates was a necessary condition but might not have been sufficient alone. The same is true in reverse: NGO critique without the existence of mediators inside the organization would not have been sufficient to trigger change. In this respect, staff composition is important. Existing knowledge and experience about unsuccessful policies are not enough. There was evidence that poverty was not reduced throughout the 1980s with SAL. But new knowledge that challenges old paradigms needs to resonate among relevant actors. Paradigms change when relevant actors give new meaning to previous ways of thinking. In the context of the World Bank, it required staff willing to break with the entrenched technocratic and econocentric culture in the Bank and able to understand the critics' concerns and translate them into Bank language. Here, dissatisfied noneconomists turned external critique into internal action. To quote one interviewee, "There is a battle going on. And we go external!" (62) (referring to another of their strategies). (63)
The new notion that environmental and social policies are not detrimental to the Bank's main objective of economic growth grew increasingly popular inside the organization. Apart from the advocates' success in formulating several operational social policies and manuals in the 1980s, the organizational structure changed and a new department was created in 1987, the Environment Department. More and more social scientists were hired by the Bank, and in 1993 their institutional position was strengthened even more by the establishment of a vice-presidency for Environmentally Sustainable Development (ESDVP). At the same time, the organization's environment had changed, reflected most significantly in the 1995 World Summit for Social Development in Copenhagen. This summit not only signified a new commitment to social development, but it also embraced a broader concept of social development as a "political, economic, ethical and spiritual vision [that is] based on human dignity, human rights, equality, respect, peace, democracy, mutual responsibility and cooperation, and full respect for the various religious and ethical values and cultural backgrounds of people." (64) This approach stresses people-centeredness, equity, social justice, and solidarity as well as the integration of social, economic, and cultural policies. At the conclusion of the summit, all participating heads of state and governments adopted the Declaration on Social Development, pledging to follow its principles.
Yet the Bank was not at the forefront of this agenda. The summit was a reflection of a changed atmosphere within the aid community. Another example was the human development approach of the UN Development Programme (UNDP), promoted by the Human Development Report Office under the leadership of Mahbub ul Haq. (65) Intellectually, the idea that development is primarily about human beings had been supported by Amartya Sen, who delivered lectures and distributed publications on the subject to Bank staff. (66) "Social" staff within the organization drew authority and legitimacy from these events. It was in this environment, and possibly a reflection of it, that James D. Wolfensohn assumed his new position as Bank president. He found an organization in crisis. The Bank was under pressure from many sides. In addition to increasing NGO critique concerning its neoliberal economic policies, which culminated in the "Fifty Years Is Enough" campaign in 1994, (67) the organization had to defend itself against crucial criticism of its "culture of loan approval." The Wapenhans Report, released in 1992, revealed that this culture of loan approval in the Bank caused a decline in the performance and quality of Bank operations. (68) These deeply rooted organizational problems were compounded by increasing financial pressure. In the early 1990s, private capital flows to developing countries rose (69) and undermined the Bank's role and influence. This, combined with the growing perception of a gap between the Bank's rhetoric and its performance, resulted in a loss of faith in aid and development.
Thus, Wolfensohn saw himself confronted with a number of challenges from diverse directions. But instead of choosing a few select priorities for the organization, he tried to be "all things to all people" in an effort to ensure the Bank's relevance, influence, and position. (70) From his first day in office, Wolfensohn promised wide-ranging internal reforms. In addition, social staff started lobbying him immediately after he assumed the presidency in 1996. (71) His first undertaking was the Strategic Compact (SC), which aimed to renew the Bank's development paradigm, its role as the leader in development, and its external legitimacy. It had a tremendous impact on the design of the social development approach, since the social agenda was at the core of the new Bank mission. In conjunction with the compact and at the urging of the Bank's lead sociologist, Michael Cernea, a social development task force was established to take stock and prepare an overall social development approach. (72) The resulting report summarized the various social policy operations already under way and provided a list of recommendations to advance the social agenda. (73)
Following these recommendations, several institutional changes were made. In January 1997, the Social Development Network was established and placed within the ESDVP; in March 1997, the SC was approved, allocating resources to social development--for example, for the recruitment of more social scientists. (74) In December 1997, a research initiative was agreed upon to work on social issues, together with the Poverty Reduction and Economic Management (PREM) unit. Moreover, by 1997, Social Development units had been created in the regions and a Social Development Board had been established with representatives from the regions. The board sought to provide infrastructure for the network by integrating and mainstreaming topics such as participation, social analysis, and gender into Bank operations; identifying key social problems in each country and region; (75) incorporating social development into the Bank's business; working with NGOs; and focusing on new topics such as postconflict reconstruction and cultural heritage. (76) Rather than scattering social development experts all over the organization, the board and the Social Development Network provided them with an institutional home, which facilitated targeted work and agenda setting.
Without the ambitious figure of James D. Wolfensohn, the Bank's reforms by the end of the 1990s might not have gone so far. However, without internal advocates lobbying the president to implement new policies, he would not have been able to exert these reforms. As Anthony Bebbington et al. point out, the World Bank can be considered a "battlefield of knowledge with different arenas in which the contests are waged." (77) NESSies are still in the minority within the Bank, and many staff still believe in economic growth as the main objective of development, although many of them have now refined their arguments. (78) Nevertheless, the existing social policy units and networks supported Wolfensohn's endeavor to raise their operational importance. Willing staff were available for the establishment of the Social Development Network, and more were hired in the late 1990s following the task force's report and the SC's suggestion. (79) This in turn increased the staff mix and thus the potential for internal lobbying to enhance the Bank's social agenda. Moreover, the SC contributed to two other change-conducive reforms that may have long-term impact: first, it required collaboration between economists and noneconomists; and, second, it changed the Bank's organizational structure into a so-called matrix structure. Both reforms led to greater competition among staff in different units and therefore to an increased degree of activity and creativity regarding new policies, which in turn might make the social agenda more prominent in Bank operations in the future.
The evolution of the World Bank's social development approach provided empirical and theoretical information. First, tracing the changes in the meaning of social (development) policies over almost twenty-five years in Bank discourse provided a useful account of the organization's understanding of development at different times and of the different roles social policies have played within the economic growth-poverty reduction nexus. The means and ends in the triad of poverty reduction, economic growth, and social policies have changed and reached a stage where poverty reduction seemed to be perceived as a prerequisite for achieving economic growth. Twenty years ago this relationship was depicted the other way around. The role of social policies has moved accordingly. "While poverty has remained, it has been framed differently over time. Thus, in this article I have described how a specific policy problem has been constructed over time, which is important to understanding specific policy responses and subsequent outcomes.
Second, in this article I have tried to discover whether the driving forces of policy change in the Bank are external or internal. The case study revealed that neither external pressure by the organization's principals nor solely the organization's internal processes explain changes to the Bank's social policy. First, member states are not the organization's sole environment. As we have seen, other actors, such as NGOs and other IOs, put significant pressure on the organization. Second, external pressure and the organization's internal conditions are intertwined. More precisely, the study revealed that agency is an important factor in triggering policy change. Internal advocates facilitate policy change in IOs by strategic use of external pressure. In doing so, they shape the content of that change. Thus, change might be different in other IOs with a different staff mixture. In the IMF, for instance--which compared with the Bank is intellectually more monolithic--change is rarely initiated internally but requires a much more powerful external event to shake the organization's self-understanding. (80) The composition of staff is a decisive factor for policy change in IOs. However, as the Bank has also shown, internal advocates are not always successful. Specific timing, sequencing, and a certain combination of events are necessary to open up policy space that allows for paradigmatic shifts to happen. Having said that, internal advocates might not be a sufficient precondition for policy change, but they are necessary.
Antje Vetterlein is lecturer in sociology at the University of Essex. Formerly she was departmental lecturer in comparative social policy at the University of Oxford. Her research focuses on international political sociology with particular interest in international organizations, global development, the linkages between processes of globalization and social policy, and the role of ideas within political discourses.
1. This is often referred to as a shift from the Washington to the post-Washington Consensus. I do not use these terms since they often have a normative connotation. For a detailed description, see Antje Vetterlein, The Politics of Development Discourse: From the Washington to the Post-Washington Consensus, PhD diss., European University Institute, Florence, 2006; or Ziya Onis and Fikret Senses, "Rethinking the Emerging Post-Washington Consensus," Development and Change 36, no. 2 (2005): 263-290.
2. World Bank, Empowering People by Transforming Institutions: Social Development in World Bank Operation: A Social Development Strategy (Washington, DC: World Bank, 2005).
3. See Michael Barnett and Martha Finnemore, "The Politics, Power and Pathologies of International Organizations," International Organization 53, no. 4 (1999): 707. See also Catherine Weaver in this issue.
4. Alexander L. George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences (Cambridge, MA: Belfer Center for Science and International Affairs, 2005).
5. On NGO influence, see, for example, R. J. O'Brien, A. M. Goetz, and M. A. Williams, Contesting Global Governance: Multilateral Economic Institutions and Global Social Movements (Cambridge: Cambridge University Press, 2000), who coined the term "complex multilateralism" to address these developments.
6. World Bank, Social Development in World Bank Operations: Results and Way Forward, discussion draft (Washington, DC: World Bank, 2004), p. 1 (emphasis in original).
8. World Bank, Empowering People, p. 1.
9. Gloria Davis, A History of the Social Development Network in the World Bank, 1973-2000, SDV Paper 56 (Washington, DC: World Bank, 2004).
10. Ibid., p. 1.
11. OED, An OED Review of Social Development in Bank Activities (Washington, DC: World Bank, 2004). Referring to "social" staff includes staff in other "social" units, such as Social Protection, which was the first to have a Bankwide SSP.
12. I am only depicting Bank-internal discourse on the relationship between the social and poverty rather than offering an academic opinion. Sometimes social development seems to be equated with poverty reduction in Bank discourse.
13. World Bank, World Development Report: Poverty and Human Development (Washington, DC: World Bank, 1980), p. iii.
14. Ibid., p. 35.
16. Lionel Demery and Tony Addison, The Alleviation of Poverty Under Structural Adjustment (Washington, DC: World Bank, 1987), p. 5.
18. World Bank and IMF, "Adjustment with Growth: The Fund, the Bank, and the Country Experiences," Finance and Development (June 1987): 7.
19. See Demery and Addison, Alleviation of Poverty; or World Bank and IMF, "Adjustment with Growth."
20. Norman L. Hicks, Poverty, Social Sector Development and the Role of the World Bank, HRO Working Paper (Washington, DC: World Bank, 1993).
21. World Bank, World Development Report 1990 (Washington, DC: World Bank, 1990).
22. Barber Conable, "Address to the Board of Governors," in World Bank, ed., Summary Proceedings: 1990 Annual Meeting of the Board of Governors (Washington, DC: World Bank, 1990), pp. 12-20.
23. World Bank, World Development Report 1990, p. 4.
24. World Bank, Economic Advisory Staff, Poverty Reduction and Bank Operations, Report No. 8491 (Washington, DC: World Bank, 1990), p. 3.
25. World Bank, Assistance Strategies to Reduce Poverty, World Bank Policy Paper (Washington, DC: World Bank, 1991), p. 12.
26. Nancy Birdsall, Social Development Is Economic Development, WPS 1123, Policy Research Department (Washington, DC: World Bank, 1993).
27. Ibid., p. 1.
29. World Bank, World Development Report 2000/01: Attacking Poverty (Oxford: Oxford University Press, 2001).
30. Ibid., p. v.
31. World Bank, New Paths to Social Development: Community and Global Networks in Action (Washington, DC: World Bank, 2000), p. 1.
32. World Bank, Social Development Update: Making Development More Inclusive and Effective, SDD Paper Series, No. 27 (Washington, DC: World Bank, 1998).
33. Memorandum, Ernest Stem to M. Syeduz-Zaman, 17 May 1982, cited in Devesh Kapur, John P. Lewis, and Richard Webb, eds., The World Bank: Its First Half Century (Washington, DC: World Bank, 1997), p. 334.
34. The Annual Reports stopped discussing poverty from 1983 until 1986.
35. Elliott Berg et al., Accelerating Development in Sub-Saharan Africa (The Berg Report) (Washington, DC: World Bank, 1981).
36. In that sense, poverty programs were regarded as unhelpful to careers (Kapur, Lewis, and Webb, The World Bank, p. 350).
37. In 1985, only 1 percent of Bank reports mention "poverty" in their title or abstract, compared with 6 percent in 1980 and 12 percent in 1995; see Robert H. Wade, "Greening the Bank: The Struggle over the Environment, 1970-1995," in Kapur, Lewis, and Webb, The World Bank, pp. 611-734.
38. Article 1 (1), available at http://web.worldbank.org/WBSITE/EXTERNAL/EXTABOUTUS/0,,contentMDK:20049563~pagePK:43912~menuPK:58863~piPK:36602,00.html#I1.
39. Kapur, Lewis, and Webb, The World Bank.
40. See ibid., p. 234.
42. Interview with Michael Cernea, 8 April 2004.
44. Kapur, Lewis, and Webb, The World Bank, p. 241.
45. World Bank, Focus on Poverty (Washington, DC: World Bank, 1982).
46. See Kapur, Lewis, and Webb, The World Bank, p. 338.
47. The Annual Reports' statistics for sector lending during the 1980s show that in the case of education, for instance, lending volume is on average about 4.5 percent of total lending (for the period 1980-1989) while for 1982-1984, the numbers were a bit below this average, and in 1987 they dropped significantly (2.5 percent); in 1981 and 1985, respectively, lending volume amounted to 6 percent and more.
48. See Davis, Social Development Network; and OED, Review of Social Development, p. 67.
49. Giovanni Andrea Cornia, Richard Jolly, and Frances Stewart, Adjustment with a Human Face, 2 vols. (Oxford: Oxford University Press, 1987).
50. See, for example, Jonathan A. Fox, "When Does Reform Policy Influence Practice? Lessons from the Bankwide Resettlement Review," in Jonathan A. Fox and L. David Brown, eds., The Struggle for Accountability: The World Bank, NGOs and Grassroots Movements (Cambridge: MIT Press, 1998), pp. 303-344.
52. See Kapur, Lewis, and Webb, The World Bank, p. 375.
53. Several interviews with anonymous World Bank officials, World Bank and IMF headquarters, Washington, DC, March 2004.
54. Interviews with Michael Cernea, World Bank headquarters, Washington, DC. 7 and 8 April 2004.
55. Michael Cernea, "Culture? ... at the World Bank? Letter to a Friend," manuscript, 2004, p. 3. 'ToR' stands for "terms of reference."
56. Two of the most significant books are Michael Cernea, Putting People First (Washington, DC: World Bank, 1985); and Lawrence Salmen, Listen to the People (Washington, DC: World Bank, 1987).
57. Interview with anonymous World Bank official, IMF headquarters, Washington, DC, 8 April 2004.
58. Cernea, "Culture?" p. 8.
59. Several interviews conducted in March/April 2004. See also Davis, Social Development Network.
60. Interviews with Michael Cernea, World Bank and IMF headquarters, Washington, DC, 8 April 2004.
61. Among many, see Cernea, Putting People First; and Salmen, Listen to the People.
62. Interview with anonymous World Bank official, IMF headquarters, Washington, DC, 8 April 2004.
63. A comparison with the IMF confirms this result. Staff homogeneity increases loyalty among staff and thus makes the organization more resistant to external pressure; see Vetterlein, Politics of Development Discourse.
64. World Bank, Social Development Update, p. 9.
65. See, for example, Louis Emmerji, Richard Jolly, and Thomas G. Weiss, "Economic and Social Thinking at the UN in Historical Perspective," Development and Change 36, no. 2 (2005): 211-235.
66. For his ideas, see Amartya K. Sen, Development as Freedom (Oxford: Oxford University Press, 2001).
67. K. Danaher, 50 Years Is Enough: The Case Against the World Bank and the International Monetary Fund (Boston: South End Press, 1994).
68. W. A. Wapenhans et al., The Wapenhans Report, Portfolio Management Task Force (Washington, DC: World Bank, 1992).
69. Private capital flows increased from $40.9 billion in 1990 to $256 billion in 1997--see Multilateral Investment Guarantee Agency, Annual Report (Washington, DC: World Bank, 1998)--whereas multi- and bilateral aid decreased from 57 percent of all net financial flows to developing countries to only 15 percent in the same time span; see World Bank, Global Development Finance, Analysis and Summary Tables (Washington, DC: World Bank, 1998).
70. Bruce Rich, "The World Bank Under James Wolfensohn," in Jonathan R. Pincus and Jeffrey A. Winters, eds., Reinventing the World Bank (Ithaca: Cornell University Press, 2002), p. 26.
71. Interviews with Bank staff in March 2004.
72. Interview with Michael Cernea, 8 April 2004. See also Anthony Bebbington, Scott Guggenheim, Elizabeth Olson, and Michael Woolcock, "Exploring Social Capital Debates at the World Bank," Journal of Development Studies 40, no. 5 (2004): 33-64.
73. World Bank, Social Development and the Results on the Ground: Task Group Report, Social Development Papers, No. 22 (Washington, DC: World Bank, 1997).
74. The SC reallocated $10 million to the regions and $2 million to ESDVP for these issues.
75. By 1997, for the first time in Bank history, each of the regions had a systematic action plan regarding social issues.
76. World Bank, Social Development Update.
77. Bebbington et al., "Exploring Social Capital Debates," p. 38.
78. See World Bank, Globalization, Growth and Poverty: Building an Inclusive World (New York: Oxford University Press, 2001). One can distinguish between "hardcore" or pure economists, soft economists, and noneconomists; see Vetterlein, Politics of Development Discourse; and Ravi Kanbur, "Economic Policy Distribution and Poverty: The Nature of Disagreements," World Development 29, no. 6 (2001): 1083-1094.
79. In 1998, 75 percent of the 220 staff mapped to the Social Development family indicated that they had been at the Bank for less than three years (Davis, Social Development Network, p. 22).
80. See Vetterlein, Politics of Development Discourse, or Antje Vetterlein, "Change in International Organisation: Innovation or Adaptation? A Comparison of the World Bank and the International Monetary Fund," in Diane Stone and Christopher Wright, eds., The World Bank and Governance (London: Routledge, 2006).