Implementation of EU development policy on the ground : a comparative study on India and Indonesia

(1)

Implementation of EU development

policy on the ground

A comparative study on India and Indonesia

Nishi Kesharie

Supervisor: Sebastian Krapohl

20 August 2015

Master thesis

(2)

2

List of abbreviations

EC European Commission

EU European Union

FDI Foreign direct investment

GDI Gender development index

GDP Gross domestic product

GII Gender inequality index

GNP Gross national product

HRW Human Rights Watch

HVW Highly vulnerable women

IBL Indonesia Business Links

MDG Millennium Development Goal

MoWF Ministry of Women Empowerment and Child Protection

NCDHR National Campaign on Dalit Human Rights

NGO Non-governmental organization

OBC Other Backward Castes

SC Scheduled Castes

SFCG Search For Common Ground

SEWA Self Employed Women’s Association

ST Scheduled Tribes

UN United Nations

(4)

4

1. Introduction

Gender equality and women’s empowerment are global issues that remain important

worldwide and it is evident that there are large contrasts in women’s lives around the globe. The United Nations (UN) distinguishes between categories of human development and emphasizes that Western and Southern Asia, North Africa and sub-Saharan Africa are areas where a focus on gender equality and women’s empowerment is necessary.1_{Scholars such as} Kabeer argue that while there may be other forms of socio-economic disparity that are wider than gender, gender inequality is more pervasive than other forms of inequality as it is a feature of social relations in most societies (Kabeer 2003: xiii). Furthermore, gender

inequality is pervasive across different groups within societies as it is a feature of rich as well as poor groups, racially dominant as well as racially subordinate groups and in privileged as well as untouchable castes (ibid., authors’ emphasis). Even though gender inequality varies across and within groups in different societies, it intersects with economic deprivation to produce more intensified forms of poverty for women than for men, which means that gender inequality is imbedded in the cause as well as the deepening of poverty in society and must therefore also be imbedded in eradicating poverty (ibid.).

Achieving gender equality and encouraging women’s empowerment is thus not solely about human rights and human development, but also a crucial component of sustainable

development. Half of the population consists out of women, and therefore women’s

development is the most effective strategy to enable the poorest and most marginalized people and communities to evolve and improve their situations (O’Connell 2014: 6). The impact of women’s development is vast, as women reinvest much more of their income in their families and communities (ibid.). Some estimates also suggest that an increase in the number of girls attending school can increase a country’s gross domestic product (GDP) and that more years of schooling in girls leads to an increase of wages by as well as better maternal health, fewer and healthier children and greater economic opportunities. 2_{Even though gender equality and} the rights of women and girls are first and foremost a matter of justice, having rights enables women to improve the situations of themselves, their families and their communities. This means that investing in girls and women can transform lives, families, communities, societies and economies. Due to the importance of this issue, as well as the major differences of gender

1_{See UNDP MDG 3 (http://www.undp.org/content/undp/en/home/mdgoverview/mdg_goals/mdg3.html)} 2_{See Juhansone, 2015 (https://europa.eu/eyd2015/en/plan-international/posts/yee-latvia)}

(5)

5

inequality and human development between states on a global scale, gender equality and the empowerment of girls and women have become important issues on the international agenda that are interconnected and integrated with poverty eradication and development in order to bring forth progress.

1.1 The EU’s role in development aid

The European Union (EU) plays an important role as a global proponent of aid as EU development aid – including aid from member states and the EC programmes – accounts for more than half of worldwide official development instances aid (Carbone 2013: 341). The EU is also generally seen as one of the pioneers in development aid, as the EC has presented itself as a benevolent actor concerned with bringing development to the rest of the world through instruments such as external aid (Dimier 2014: 1). The EU has always emphasized gender equality, as it initiated a gender policy since its beginnings in 1957, which was then based predominantly on the concept of equal opportunities (Verloo 2007: 49). The EU still recognizes the importance of gender equality and attempts to address these issues in their development policy by considering equality between women and men a key principle of European integration, and being committed to include a strong gender component in all its policies and practices in its relations with developing countries in addition to financing targeted actions to help girls and women.3

The EU has two set ways of reducing gender inequality and promoting women’s

empowerment in their development policy. First of all through the mainstreaming of gender in all their policies, which means that gender is taken into consideration in any planned action - including legislation, policies or programmes – in all areas and at all levels in order to integrate women’s as well as men’s concerns and experiences in political, economic and societal spheres (Moser and Moser 2010: 12). The EU considers gender mainstreaming as a very powerful tool as it relocates gender equality from a marginalized position into the mainstream (Brodolini 2012: 20). Gender mainstreaming is an overall strategy, whereby budget is crucial as it is the most important economic policy instrument of government (ibid.). Therefore, budget decisions that integrate a gender perspective appear to be pivotal to the application of the gender mainstreaming strategy (ibid.).

3_{See European Year for Development} (https://europa.eu/eyd2015/en/eu-european-parliament/posts/economic-and-political-discrimination-undermines-womens-opportunity)

(6)

6

Second and last of all, through a standalone focus on gender which is manifested in the execution of the Millennium Development Goals (MDG) (Kabeer and Woodruffe 2014: 24). Two of the MDGs focus specifically on women’s development, namely MDG 3 focuses on gender equality and promotion of women’s empowerment, and MDG 5 focuses on improving maternal health. Even though the MDGs are a United Nations initiative, the EU formally committed to promote these values in 2000 and integrated the third MDG of promoting gender equality and empowering women into its development policy and practice.

1.2 Effectiveness of EU development aid policy

There has been some success regarding the EU’s development aid policies as the political participation of women has increased, and the world has achieved equality in primary education between girls and boys4. However, there is also a widespread perception that EU development aid has not been effective as most target sets of including gender in policies are not achieved, gender gaps in education and employment remain and Millennium

Development Goal (MDG) targets are missed (Menocal et al. 2007: 3). The ineffectiveness of foreign aid has led to a new agenda of ambitious reforms in in the international aid system such as the 2005 Paris Declaration on Aid Effectiveness which is supposed to increase policy effectiveness, increase economic growth and reduce poverty on the ground. However,

scholars express explicit criticism due to the lack of the integration of human rights in EU development policies (Foresti, Booth and O’Neil 2006: vi)

1.3 India and Indonesia

This difference in effectiveness of EU development policy is also perceptible when looking at the current situation in Indonesia and India with regard to gender equality and women’s empowerment. India and Indonesia are both upcoming economies that are identified by Goldman Sachs as having a high potential of becoming the world’s largest economy in the 21st century. In 2001, Goldman Sachs economist Jim O’Neill argued that the BRIC countries – Brazil, Russia, India and China – would show significant worldwide growth in relation to the developed G7 economies in the next few years (O’Neill 2001: 1). He predicted that India and China would become the dominant global suppliers of manufactured goods and services, increasing their economic growth and therefore be able to replace two states within the G7 (Barker 2013: 1). Since 2003, India has become one of the fastest-growing major economies due to increases in per capita income and integration within the global economy (Poddar and

(7)

7

Yi 2007a: 11). Moreover, India’s growth acceleration since 2003 is considered to represent a structural increase rather than a simple cyclical upturn (Poddar and Yi 2007b: 3).

Whereas India is one of the BRIC countries, Indonesia is one of the MIST nations. In 2011, O’Neill coined the term and claimed that the so-called MIST nations – Mexico, Indonesia, South Korea and Turkey – are currently growing faster than the BRIC countries, have a major effect on the global economy and are therefore regarded as the new biggest emerging markets (Yilmaz and Hussain 2012: 9). Indonesia is considered the leading economy of Southeast Asia and has been displaying unexpected rapid growth (ibid.: 40).

Both countries play an important global and regional geostrategic role as upcoming economies. Simultaneously, both countries are still dealing with poverty, medium human development and gender inequality. They have other similarities, such as strong colonial ties to the Netherlands and the United Kingdom respectively, similar political regimes (although Indonesia is sometimes considered a flawed democracy while India is sometimes seen as a strong democracy) and religious populations. However, Indonesia is doing considerably better with regard to gender equality, women’s rights and women’s empowerment than India, as women in Indonesia have relatively more equal access to education than women in India, as well as a longer and healthy life and a decent standard of living.5

This discrepancy between policy effectiveness in Indonesia and India may have several different causes, as both countries consist out of complex, multi-faceted societies and as the EU has different objectives in their relations with these countries. Foreign aid effectiveness is often linked to good governance as well as to conditionality and policy implementation (Foresti, Booth and O’Neil 2006: 12). Policy implementation is especially valuable to take into consideration when studying effectiveness, as policy change may not lead to desired results if the process of implementation is omitted from consideration (Cema 2013: 3).

In this thesis, I will look closer to the relationship between the implementation of EU development policy with regard to gender equality and women’s empowerment and the effectiveness of this policy and its objectives in India and Indonesia. The research question will therefore be: Why is nationally implemented EU development policy more effective in Indonesia than in India with regard to gender equality and women’s empowerment.

(8)

8

1.4 Relevance of the thesis

The relevance of this thesis mostly lies in the importance of women’s empowerment and gender equality on its own, as well as in relation to poverty eradication and development. It is a matter that is worth studying on its own, but is also important to analyze as a part of

development studies in order to generate knowledge on bringing forth progress. Furthermore, understanding if and why the policies we analyze are effective has the potential to provide an important contribution to policy research, because policy implementation and policy

outcomes are integral parts of the policymaking process and must be studied to obtain an accurate view of a policy (Stoutenborough & Oxley 2012: 3). By studying the effectivity and implementation of policy in Indonesia and India, we can gather insights in these specific cases as well as garner more general knowledge on implementation of development aid policy with regard to women’s empowerment and gender equality in other nations.

(9)

9

2. Modernization Theory

2.1 Introduction

Modernization theory focuses on the transition from traditional, pre-industrialized society to modern, industrialized society through models of development and economic growth (Peet and Hartwick 2009: 121). In the 1950s and 1960s, ideas behind modernization theory were popularized by authors such as Walt Rostow. These modernization theorists expected economic growth to facilitate basic social protections, thereby increasing the urban middle classes and laying the social foundations for the consolidation of democratic institutions and civic society (Inglehart & Norris 2003: 4). It was therefore implicitly assumed that endemic problems of gender inequality, such as women’s literacy and education, women’s poverty, low pay and segregation in the workforce and participation and representation in the political system would be solved as well (ibid.: 4 – 5). However, by the end of the 20th_century, scholars noted that economic growth has its limitations as wealthy states did not achieve an increase in gender equality. Inglehart and Norris explain that some states in the Middle East have become as rich as some European states in per capita GDP, but most women in richer Middle Eastern states continue to have narrowly restricted rights and opportunities in the public domain (ibid.). Whereas economic growth does generally lead to more literacy and education and social security, women do not automatically benefit from this growth and therefore problems of gender equality are more complex than early developmental theorists assumed (ibid.).

2.2 Criticizing modernization theory

The academic community has critiqued original modernization theory because it assumes that all developing societies are homogenous and will evolve linearly, and because it ignores relations between societies and the ways societies influence each other as it assumes that all societies will evolve parallel to each other (Inglehart 1997b: 449). Modernization theory is also often seen as ethnocentric and Eurocentric in the sense that it takes the western society as its model, does not recognize non-western history of traditional societies and therefore

(10)

10

2.3 Revised modernization theory

Even though the notion that economic development will automatically lead to gender equality in poorer societies still remains, a large part of the scholarly community thus sees the

limitations of modernization theory (ibid.: 5). Therefore, Inglehart and Norris use a revised version of modernization theory to argue that economic growth and development brings forth human development and changed cultural attitudes towards gender equality in societies that experience forms of modernization (Inglehart and Norris 2003: 11-12). Modernization theories from the 1950s, 1960s and 1970s argued that economic, cultural, and political changes go together in coherent ways and that industrialization brings forth broadly similar trajectories in different societies (ibid.). However, Inglehart and Norris argue that whereas economic growth increases change, the changes are probabilistic instead of deterministic (ibid.: 11). Inglehart and Norris’ argument thus does not subscribe to the idea that all developing societies are homogenous which will develop linearly and does not have the Western state as its model.

Inglehart and Norris state that human development brings changed cultural attitudes towards gender equality in any society that experience the various forms of modernization linked with economic development (ibid.: 10). They argue that modernization impacts this changed cultural attitude in two phases of industrialization and post-industrialization. In the first phase of industrialization, women enter the paid workforce which leads to dramatically lower fertility rates. Women attain literacy and greater educational opportunities and begin to participate in representative government even though they still have far less power than men. After industrialization occurs, the post-industrial phase causes women to rise in management and gain political influence within elected and appointed organizations, therefore creating a shift towards greater gender equality.

(11)

11

According to Inglehart and Norris, only the more advanced industrial societies are currently moving towards this trajectory as over half of the world has not entered this phase yet. In their theory, these two phases of industrialization and post-industrialization are linked to two dimensions of cross-cultural variation. First of all, there is a transition from what are called “traditional” values to secular-rational values. In this phase, the traditional family declines. Second of all, there is a transition from survival to self-expression values, in which gender equality will arise. It is emphasized that cultural shifts in modern societies are not solely responsible for gender equality, but rather lay the basis for structural reforms and women’s rights and therefore facilitate the process. The transformation in postindustrial societies of a focus on material concerns to quality-of-life values that include gender equality is thus related to socioeconomic changes. Modernization theory according to Inglehart and Norris sees economic, political and cultural changes evolving together in coherent trajectories without claiming that change is either driven solely by economic factors or solely by cultural factors (ibid.: 11 – 15).

2.4 Hypothesis

Based on Inglehart and Norris’ revised modernization theory, we can hypothesize that development policy is more effective in states with higher economic development, as

Industrialization

Post-industrialization

Secular-rational

values

Traditional

values

(12)

12

economic development leads to modernization and a focus on quality-of-life values. We expect that societies that experience modernization will be more susceptible to the

implementation of development policies by the EU because there is less internal resistance towards gender equality due to higher pursuit of self-expression values. The hypothesis derived (H1) is as follows: Nationally implemented development policy is more effective in countries with higher economic development because they experience forms of modernization.

(13)

13

3. Theorizing Islam, Hinduism and the role of the caste system

3.1 Introduction

One of the major differences between Indonesia and India lies in their culture and religion. Indonesia has the largest population of the world’s Muslims, whereas India is a Hindu

majority country. Islam is sometimes seen as a civilization lacking gender equality. However, that generalization is often made with respect to the Arab world as the non-Arab Muslim world has a distinctively different Islamic culture. Hinduism’s main trope is the caste system, which is seen as a hierarchical system that impedes the general improvement of poor, lower caste, peoples. It is thus important to theorize the effects that (non-Arabic) Islam and

Hinduism have on the development of gender equality and women’s empowerment. Based on the literature, I will argue that the caste system reinforces gender hierarchies and therefore is an obstacle to gender equality.

3.2 Islam, gender and democracy

Justice and equality are intrinsic values in the Islam and the sharia. However, justice and equality are not always reflected in gender relations and the rights of men versus the rights of women in Muslim society (Mir-Hosseini 2006: 629). Therefore, the Muslim world is often depicted as a homogenous civilization lacking democracy and gender equality (Spierings, Smits and Verloo 2009: 503). Since Samuel Huntington's controversial ‘clash of civilizations’ thesis, scholars have increasingly studied the relationship between Muslim countries and the West and their differences and view democracy as an important issue of contention. Some authors claim that Islam and the West claim due to the Islam’s incompatibility with open government (Fish 2002: 4), whereas others argue that citizens in Muslim countries have significantly less tolerant attitudes towards equal rights and opportunities for women which creates an incompatibility with democracy, as gender equality is one of the core

characteristics of democracy (Inglehart and Norris 2003: 65). The relationship between gender and democracy is important, as the quality of democracy is partly determined by the extent that different social groups participate in democratic institutions (Moghadam 2004: 2). Therefore, the absence of women from the political life in a society results in democratization that is male dominated and therefore incomplete and biased (Rizzo et al. 2007: 1153,

Moghadam 2004: 2).

Fish (2002) examined the relationship between Islam and regime type and claims that Muslim countries are democratic underachievers due to the subordination of women (Fish 2002: 4 - 5). He states that Muslim societies are distinct from other societies in the treatment and status

(14)

14

of women and girls (ibid.: 24) This may affect politics as the repressiveness and unquestioned dominance of the male in relations between men and women replicate themselves in broader society, creating a culture of domination, intolerance and dependency in social and political life (ibid.: 24, 30). Furthermore, the social marginalization of women may remove distinctive voices and influences from politics, affecting regime types (ibid.: 31).

However, Rizzo, Abdel-Latif and Meyer state that in order to draw conclusions on gender equality, democracy and the Muslim world, it is essential to examine differences in gender equality and democracy within the Muslim world as the Muslim world is not a homogenous entity and should not be studied as one (2007: 1153 – 1154, author’s emphasis). The need for a nuanced examination of Muslim societies is necessary because the two subsets of countries share Islamic religious traditions, but differ dramatically in electoral competitiveness and thus the nature of Islam by itself cannot explain the lack of democratization in the Arab Muslim world (ibid.: 1154).

Studying Arab countries in comparison to non-Arab countries has pointed out that most Arab Muslim nations are democratic underachievers, whereas a significant amount of Muslim majority non-Arab countries are democratic overachievers and are also generally more electorally competitive than Arab Muslim majority countries (ibid: 1153 – 1154). Scholars have also emphasized the presence or absence of women in high political positions in Arab or non-Arab Muslim societies, which has shown that countries that are part of the ‘classic patriarchy’ that extends from North Africa, through the Muslim Middle East to South and East Asia is characterized by the low status of women in general as a result of patrilineal-patrilocal households, high fertility rates and low age at first marriage, high maternal and infant mortality rates, higher rates of female illiteracy and lower levels of female educational enrollment, low female labor force participation, and the lack of women's political

participation and political rights (ibid.: 1154). Additionally, there has never been a female head of state in the Arab Muslim world, while there have been female Presidents or Prime Ministers in non-Arab Muslim6_{(ibid.). Even though the majority of the women in these} positions enjoyed privileges that increased their chances of attaining these positions due to their connections to political families where their husbands or fathers also held high positions of power, Rizzo et al. conjecture that the visibility of female political leaders affects the sentiments of ordinary citizens regarding gender equality and democracy (ibid. 1154).

6_{According to the 2000 World Values Survey, Indonesia, Bangladesh, Pakistan and Turkey had female} Presidents or Prime Ministers.

(15)

15

Rizzo et al. also examined how support for gender equality affects support for democracy by comparing Arab Muslim countries to non-Arab Muslim countries on the issues of gender equality and democratization. They found that the majority generally did not support gender equality on issues such as equal rights to jobs and political leadership, and believed that a wife must always obey her husband (ibid.: 1162). However, the minority in the non-Arab Muslim world who did support gender equality on those issues was significantly larger than the minority in the Arab countries (ibid.). Rizzo et el. conclude that non-Arab Muslim

populations had attitudes that are more compatible with democratic forms of government, were more skeptical of religious leaders providing answers to social problems, and larger minorities were more critical of democracy even though the majority feels it is the best form of government (ibid.).They also concluded that there were higher levels of support for

women’s rights in non-Arab Muslim societies, and those who supported gender equality were much more likely to support democracy whereas the reverse was true for Arab Muslim societies (ibid.: 1165). Drawing on these findings, we can state that Muslims living in Arab countries tend to think differently from those in non-Arab countries and that gender equality and democracy are interconnected as gender equality must be supported in order for

democratization to be complete and unbiased (ibid.: 1166).

Spierings et al. don't just make the distinction between Arab and non-Arab Muslum societies, but divide the Muslim world into four regions according to their historic-cultural background: the Middle Eastern region, the Sub-Saharan region, the former Soviet region and the

Southeastern Asian region (2009: 506). They argue that the degree to which orthodox Islam is integrated into the state structure, and the level of democracy are important in explaining gender equality (in the labor market), as well as cultural and social heritage and the presence of oil resources (ibid.: 504). They find that there are strong variations in the participation of women on the labor market in the four regions, but ultimately conclude that strong

incorporation of Islam in the state structure does not affect women's participation in the formal economy, and that the absolute number of women participating on the labor market is primarily due to economic factors (ibid.: 513-514, 518). They thereby refute the idea that state Islamization is negatively related to women's economic independence or economic power, even though they state that the political opportunity structure of democracy and Islamization are important factors (ibid.: 518).

(16)

16

3.3 Hinduism and the caste system

In its classical articulation, the caste system is considered the backbone of traditional Brahmanical Hindu society. The caste system of pre-British Hindu society was a fluid, non-rigid system that allowed social mobility, where people were born into occupations but not obliged to fulfill those occupations in their entire life (Srinivas 1968: 189). Under the British influence, the caste system became more rigid, even though the British created some new avenues to mobility (ibid.: 191). The caste system was abolished during India's independence in 1947, but the political culture of caste continues to play an important role in India

(Sathyamurthy 1996: 44). The three thousand year old system has been analyzed by anthropologists as “a sacralized social order based on the notion of ritual purity” (Jaffrelot 2000: 757), while other scholars have stated that hierarchy is the fundamental social principle that lies at the heart of the system (Dumont 1970: 2).

In English, the term caste refers to two distinct concepts in Hinduism; varna, the national system and jati, the local system (Liddle and Joshi 1986: 57). The varna system divided the ancient Hindu society into four distinct varnas (castes), that are mutually exclusive,

hereditary, endogamous and occupation specific, namely: Brahmins (priests and teachers), Kshatriya (warriors and royalty), Vaisya (traders and merchants) and Sudras (servants). A fifth varna, the Ati Sudra, was later added and refers to those doing the most menial jobs or the former untouchables (Deshpande 2000a: 383-384). Jati refers to the operative, local category that determines the contemporary social code, and is much more complex in terms of hierarchy and code of conduct (ibid.: 384). There are thousands of jatis, and a certain jati does not always correspond with a certain varna (ibid.). A jati can therefore claim a varna status without other jatis disputing this claim - this illustrates the mobility and fluidness of the system (ibid.). It also exemplifies that individual jatis and their varna counterparts are complex to unravel, and that caste divisions are not dichotomous (ibid.). The inequality resulting from the caste system is therefore often analyzed by emphasizing the sharpest contradiction of those at the top and those at the bottom of the castes, using either the jati or varna criterion (ibid.).

Some scholars who analyze Indian society turn the caste system into the central problematic issue of Indian society and place hierarchy as a concept at the heart of a sociology for India (Appadurai 1986: 745). Other scholars, such as Srinivas, illustrate in what ways the system allows for social mobility and highlight the adaptive character of the system (Srinivas 1968: 189, Patel 2005: 105, 108). Drawing on the works of Liddle and Joshi (1968), Chakravarti

(17)

17

(1993) and Deshpande (2002), the argument in this thesis suggests that the caste system is a crucial hindrance to gender inequality and the empowerment of women and that it is partly because of the caste system’s fluidity and possibilities of social mobility, that gender hierarchy is continuously reinforced.

3.4 Gender and caste

According to Liddle and Joshi (1986) and Chakravarti (1993), gender and the caste system are interconnected as the development of the gender division and hierarchy was integral to the formation of the social brahmanical structure of the system (Liddle and Joshi 1986: 57, Chakravarti 1993: 579). One of the cultural aspects of the caste system is the scale of (ritual) purity and pollution, and increased control over women is one of the factors that a caste must observe (among others) before it can claim to be ritually pure (ibid.: 58 – 59). Therefore, the lower caste’s failure to control women was partly what made them impure (ibid.: 69). Higher castes control women by excluding them from the productive economy, involving removing them from the public sphere to the domestic sphere in practices of seclusion, as well as control over women’s sexuality through arranged marriage, child marriage, the prohibition of divorce and strict monogamy for women (ibid.: 59). Lower casts that improve their

economic position could attempt to move up the hierarchy over generations through processes of Sanskritization (Deshpande 2002: 25). In addition to increasing economic power the caste also has to adopt cultural attributes of ritual purity, which means constraining women’s freedom (Liddle and Joshi 1968: 59).

As the emulation of practices of higher casts thus increases purity, the lower castes attempted to behave like higher castes in order for the distinctions of caste to dissolve (ibid.: 69). However, both attempting to control women as well emulating higher castes reinforces caste division, which means that gender divisions reinforce caste divisions and that the gender ideology of controlling women legitimates the structure of patriarchy as well as the organization of caste (69).

Additionally, Chakravarti argues that the structure of social relations which shaped gender in the caste system was reproduced and reinforced by achieving the compliance of women (1993: 585). The values of the caste system were accepted by both men and women of the upper castes, through a combination of consent and coercion, as the social structure rewarded women at the same time as it subordinated them (ibid.). Women internalized patriarchal values by attempting to live up to the ideal notion of womanhood constructed patriarchal

(18)

18

ideologues (ibid.: 582). When they were not successful, they were coerced (ibid.: 583). According to Chakravarti, the reinforcement of gender hierarchy in the caste system through women’s internalization is thus subtle, and therefore very successful.

3.5 Dalit7_{women as ‘the worst off’}

As Liddle and Joshi stated, higher castes subjected women to increasing constraints and thus the subordination of women was crucial to the development of caste hierarchy (Deshpande 2002: 25). This means that upper castes relegated less autonomy to women, whereas lower castes (or Dalit castes) have historically been considered relatively more egalitarian for women (Deshpande 2002: 25). The literature therefore debates if the relative deprivation of Dalit women compensates for their relative autonomy, but Deshpande establishes in her study that overall, Dalit women are the worst off (2002: 19). Present-day caste inequalities are not just remains of the past, but continue to be perpetuated and reinforced in the present (ibid.: 20).

First of all, Deshpande’s study points out that with regard to decision-making and freedom of movement, Dalit women enjoy marginally higher levels of autonomy compared to non-untouchable castes and have relatively egalitarian spousal relations with regard to family planning (ibid.: 27-28). Figures on domestic violence show that Dalit women reported

significantly more physical violence. As the prevalence of domestic violence decreases as the standard of living increases, Deshpande argues that domestic violence is greater among Dalit women than upper-caste women. Additionally, lower castes emulate upper-caste practices as symbolic of a betterment in their position through Sanskritization (ibid.: 25). An important aspect of Sanskritization is the emulation of immuring women, which therefore leads to less relative freedom and autonomy for Dalit women and makes the distinction of autonomy between upper caste and lower caste women redundant (ibid.).

Second of all, Dalit women are continuously at the bottom of the rung across India when it comes to education, occupation, landholding, asset ownership and livestock ownership (2001: 337). Therefore, there continues to be high material deprivation of the Dalit castes and it is apparent that the women in these groups share this deprivation. The exact gap between men and women cannot be ascertained, but Deshpande states that improvement in the material condition of Dalits is not at par with other non-untouchable castes or the national average,

7_{Lower castes are officially called Scheduled Castes (SCs) by the Indian government and are commonly known} as Dalits (meaning ‘the oppressed) in Indian society.

(19)

19

meaning that in terms of material deprivation and standard of living, Dalit women are significantly worse off than upper-caste women (2002: 26-27).

Third and last of all, Deshpande establishes that patterns of low education and material deprivation often lead to further neglect of women, which alludes to a pattern that is true for all poor women in India. However, this vicious cycle may be the hardest to break for Dalit women as Dalit women endure a combination of poverty and gender discrimination that keeps them illiterate, low paid, malnourished and unhealthy (ibid.: 32). The relatively low degree of autonomy and lack of mobility leads to uneven distribution of resources in households, thereby reinforcing poverty (ibid.: 32). Deshpande finally concludes that Dalit women are the worst off in comparison to all other groups in India, as they belong to a group that is

materially at the bottom of the ladder and their relative deprivation is compounded by low levels of autonomy and greater exposure to domestic violence (2002: 19).

3.6 Hypothesis

Based on the literature, we can expect the caste system to be an additional hindrance in the implementation of gender equality policies and other development policies, due to the way the caste system has affected Indian society and continues to affect Indian society.

First of all, the subordination of women was crucial to the development of caste hierarchy. Additionally, the caste system is reinforced by gender ideologies, as lower castes emulate higher caste practices based on gender divisions. Second of all, the economic condition of women continues to be defined and constrained by their caste status. Low caste women are the worst off, as they have low levels of autonomy and greater exposure to domestic violence. They belong to a group that is at the bottom of the ladder, and it is significantly more difficult for lower caste women to move upwards as they endure a combination of poverty and gender discrimination that doesn’t allow for social mobility.

Even though Arab Muslim society generally does not support gender equality, scholars have found that non-Arab Muslim society think differently and therefore may provide a more susceptible ground for EU development policies to be effective. The hypothesis derived from these premises is as follows (H2): The caste system in Hindu society may be an additional hindrance to the implementation of gender equality policies.

(20)

20

4. Policy implementation framework

Paul Sabatier and Daniel Mazmanian identify a set of five sufficient and generally necessary conditions under which a policy decision that seeks a substantial departure from the status quo can achieve its policy objectives and thus how it can be effective (1979: 483).

The five conditions are as follows (ibid.: 485):

1. The program is based on a sound theory relating changes in target group behavior to the achievement of the desired end-state (objectives).

2. The statute (or other basic policy decision) contains unambiguous policy directives and structures the implementation process so as to maximize the likelihood that target groups will perform as desired

3. The leaders of the implementing agencies possess substantial managerial and political skill and are committed to statutory goals.

4. The program is actively supported by organized constituency groups and by a few key legislators (or the chief executive) throughout the implementation process, with the courts being neutral or supportive.

5. The relative priority of statutory objectives is not significantly undermined over time by the emergence of conflicting public policies or by changes in relevant

socioeconomic conditions that undermine the statute's "technical" theory or political support.

The authors emphasize that even if the conditions for effective implementation are not all met at the time of the basic policy decision, policy formulators and other program supporters can still take a number of steps to approximate the ideal over time and work towards effective policy implementation (ibid.: 504).

The first necessary condition under which policy can be efficiently implemented has to do with relating achievement of the desired (target) goal to changes in target group behavior, and also specifying the means by which target group compliance can be obtained (ibid.: 486). For the development policy of the EU, this means that technical components of the theory

(21)

21

empowerment and maternal health) should be operationalized in order to positively affect the achievement of policy objectives.

An example of this is the EU Development Policy. According to the European Commission, the EU Development Policy before 2000 lacked clear policy objectives, whereas after 2004, the European Commission identified a number areas requiring improvement (the desired target goal) and addressed problems of poor governance, corruption, human rights and weak government commitment to development policies in order to target group behavior (Dearden 2008: 5).

With regard to the implementation of EU women’s development policy in India and Indonesia, this means that the EU has to link underlying goals of women’s development theories – such as the influence women as decision makers have on women empowerment – to target group compliance by operationalizing policies in ways that are aimed at changing group behavior.

The second necessary condition under which policy implementation will be most effective is when the policy is straightforward, unambiguous and maximizes the likelihood that target groups will perform as desired. According to Sabatier and Mazmanian, this condition is most overlooked by policy makers. An important aspect of this condition, is that implementation is assigned to agencies supportive of the objectives that will give the policy and forthcoming programs high priority (Sabatier and Mazmanian 1979: 489). In development literature, this can point to close cooperation with civil society and non-governmental organizations, because officials that are sufficiently committed and persistent to developing new regulations and procedures and enforcing them are more likely to lead to effective and efficient results (ibid.). For this reason, Sabatier and Mazmanian argue that minimizing the number of veto points and providing hierarchical integration within implementation agencies in order to improve

obtaining coordinated action (ibid: 491). By minimizing veto power of other actors, especially in federal organizations, the reliance on state and local agencies for carrying out the details of the policy programmes will be minimized as well, making resistance of those state and local agencies less plausible (ibid.). Top down approaches suffer when local organizations work against policies, therefore also minimizing targeting group compliance and behavior (ibid.: 494). Working with civil society and local non-governmental organizations (NGOs) will result in motivated and supportive policy implementation, whereas minimizing veto power of

(22)

22

resistant federal organizations or legislators will ensure less resistance and thus more effective implementation.

The third condition states that leaders of the implementing agencies should be motivated and committed to achieving objectives, and possess substantial managerial and political skill. This condition is crucial because policy support is useless if not accompanied by political and managerial skill, which entails the ability to develop good working relationships with sovereigns, convincing opponents and target groups of the policy, and mobilizing support among civil society and non-governmental organizations (ibid.: 495).

Even though the EU development policies have not always been successful, they are globally seen as innovative policies, and according to Sabatier and Mazmanian, that should be enough to attract committed and skillful executives. Because women's development should be in close cooperation with local government, civil society and non-governmental organizations, it is important to consider that policy implementation will be most effective if executives on the ground have just as much managerial and political skill as EU legislators.

Closely affiliated with the second necessary condition that emphasizes the importance of implementation assignment to agencies supportive of the objectives that will give the policy high priority is the fourth necessary condition, which entails active support by organized constituency groups and by key legislators, as well as neutral or supportive courts (ibid.: 496). Even though political support seems easy to accomplish if the first three conditions are met, Sabatier and Mazmanian argue that it is difficult due to three reasons. First of all, short issue-attention span of the general public and mass media tend to undermine political support for particular programs. Second of all, there is a general tendency for support by agencies for certain programs to decline over time - especially development programs. Third and last of all, most legislators lack the resources and/or the incentives to monitor program

implementation actively (ibid.). This means that without active political support from key legislators, delay and resistance will increase among the public as well as resistant federal legislators (ibid.: 497). Executive officials who control resources and enough staff to closely monitor the implementation process, intervene with agency officials and protect the budget and legal authority is the answer to these problems. An organized supportive constituency (interest) group backing up the policy implementation will provide these executive officials and therefore increase policy implementation effectiveness. It is also important to take courts into account, due to their role in procedural issues and their authority to weaken

(23)

23

implementation through delay (ibid.: 499). For EU development policy, this means that the EU needs to work together closely with organized constituency groups and key legislators in developing countries. Due to the power of courts, policy implementation will be most successful if the policies that are to be implemented are not opposed to the courts.

The fifth necessary condition states that policies should not be opposed to (conflicting) public policies or by changes in socioeconomic conditions that undermine the policy's technical theory or political support. EU development policy should be closely tied to ongoing political, social and economic events in developing countries and the underlying policy theory should be based on the ongoing sociopolitical and socioeconomic situation. Because women's development is even more subjected to sociopolitical and socioeconomic change in developing societies, EU development policies should be well researched in order to withstand those changes. Due to occasional unstable or fickle changes in all societies - but especially developing societies - it is important that support from key legislators, organized constituency groups and officials are crucial in order to ensure that policy survives and is effectively implemented.

(24)

24

5. Research Methods

5.1 Comparative case study

Qualitative research focuses on a small number of cases and uses in-depth analysis to describe complex phenomena. In order to answer why nationally implemented EU development policy is more effective in Indonesia than in India with regard to gender equality and women’s empowerment, a comparative case study approach is most suitable. A case study is defined as “the intensive study of a single case for the purpose of understanding a larger class of cases (a population)” (Gerring 2007: 96). Case studies allow us to study a causal mechanism or

relationship, and locate the factors lying between some structural cause and its supposed effect and therefore allow us to ‘see’ the independent and dependent variable interact (ibid.: 102). Social phenomena are also better understood if they are compared in relation to two cases or situations, and by comparing situations the researcher is also in a better position to establish the circumstances in which a theory will or will not hold (Bryman 2008: 58 – 59).

Researchers generally distinguish between internal and external validity, where internal validity means internal to the sample under study and external validity applies to a broader - unstudied - population (Gerring 2007: 101). Case study research has problems of

representativeness because it includes only a small number of cases of a more general phenomenon, and they are therefore considered weaker with respect to external validity (ibid.). However, this also means that case study research generally has stronger internal validity (ibid.). It is also often easier to establish the veracity and therefore the validity of a causal relationship concerning a single case or a small number of cases (ibid.). For this thesis, this means that internal validity will be higher than external validity as the question of

(nationally implemented) EU development policy effectiveness specifically applies to the case of Indonesia and India. However, we can draw on the findings of this study to raise questions about EU development policy effectiveness among all states that receive official EU foreign aid.

My case study consists out of the larger context of gender equality and women’s

empowerment in both India and Indonesia, and I will explicitly focus on cases of policy implementation within India and Indonesia that will be analyzed on the basis of Sabatier and Mazmanian’s theory on effective conditions of policy implementation.

(25)

25

5.2 Most similar systems design

The Most Similar Systems Design is based on the belief that systems that are similar in as many ways as possible are optimal for research, as many experimental variables are

minimized (Przeworski and Teune 1970: 32). The level of analysis is that of the systems, and common systemic characteristics are conceived of as controlled for whereas inter-systemic differences are viewed as explanatory variables (ibid.: 33). Therefore, factors that are

common to the countries are irrelevant in determining the phenomenon that is explained and the variables that differentiates the systems can be considered as explaining variables.

Case selection

India and Indonesia are both dynamic, upcoming economies that experience significant economic growth in the last decade. Both countries have strong colonial ties to their former colonizers, similar political regimes and are important regional players in South Asia and Southeast Asia respectively. Despite their similarities, there is significantly more gender inequality in India than in Indonesia, which makes these cases suitable for the Most Similar Systems Design. Economy and religion are major inter-systemic differences and therefore the main focus in this research.

Independent and dependent variables

The research question implies that nationally implemented EU development policy leads to effective policy in Indonesia (and to a lesser extent in India). Therefore, the independent variable is nationally implemented EU development policy regarding women’s empowerment and gender equality (in Indonesia and India), and the dependent variable is effectiveness of development policy (in Indonesia and India).

5.3 Process tracing

The objective of process tracing is bringing theory closer to what really goes on in the world, in order to make scholarship more policy relevant and increase the reliability of our findings (Checkel 2005: 2). Process tracing is crucial to acquire knowledge about the processes underlying potentially transformative and dynamic phenomena. The process-tracing method attempts to identify the intervening causal process - the causal chain and causal mechanism - between an independent variable and the outcome of the dependent variable (ibid.: 5). Process tracing offers the possibility of mapping out one or more potential causal paths that are

(26)

26

(the alternative paths through which the outcome could have occurred) must be taken into account.

Process tracing directs one to trace the process in a very specific, theoretically informed way. Between the beginning (independent variable) and end (outcome of dependent variable), the researcher looks for a series of theoretically predicted intermediate steps. Process tracing thus means “to trace the operation of the causal mechanism(s) at work in a given situation”, where one carefully maps the process in order to explore the extent to which it coincides with prior, theoretically derived expectations about the workings of the mechanism (Checkel 2008: 116). The data for process tracing is qualitative in nature, and process tracing is compatible within the empiricist/positivist tradition (ibid.).

5.4 Data

Data collection for this study is based on data from the UN and World Bank, policy documents in the form of evaluation reports received by the national implementing organizations and (publicly accessible) human rights reports. The EU is funding many projects that focus on governance, environmental issues and trade in Indonesia, but out of the ongoing projects only two have an explicit focus on gender equality and women’s

empowerment. The EU has much more ongoing projects that focus on gender equality and women’s empowerment in India, but evaluation of those projects occurred through an overarching report. I have received evaluation reports of Indonesian as well as Indian based projects through (the local employees of) the Delegation of the European Union to Indonesia who have referred me to the implementing organizations. Communication with implementing organizations in Indonesia was more difficult and time-consuming than communication with implementing organizations in India8.

The first case study of Indonesia will be of the EU project entitled “A More Inclusive Democracy: Strengthening Women’s Participation for the 2014 Elections”, which is implemented by an international organization in partnership with a national Indonesian organization in three Indonesian provinces. The objectives of this project consist out of building the capacity of women running for parliament at district and municipality level in collaborative and inclusive leadership; building the capacity of political parties to foster more

8_{Indonesian implementing organizations asked for more in-depth information on my research, and because} they did not reply to all my e-mails, I had to send several follow-up e-mails compared to immediate

(27)

27

diverse leadership by supporting female candidates; and building voter and constituency support for women in politics.

The second case study of Indonesia will be of the EU project entitled “Young Women’s Economic Empowerment” and is implemented by an international organization in cooperation with four local partners. The project is suited for my case study as it has a strong gender component and a strong economic component. It aims to increase economic security for young women, by obtaining decent work or creating a sustainable business. The emphasis is on financial literacy training, in order for women to have equal access to and control over resources through microfinance.

The case study of India will consist out of ten EU funded, nationally implemented projects that are assessed by an independent team of external evaluators in an overarching policy report. Nine out of the ten projects are financed under the EC call entitled “Civil society organizations in Development: Support to in-country interventions; Vocational education and training for vulnerable and marginalized groups in India.” The other project is financed under the EC call entitled “Co-financing with European Development NGOs, Actions in Developing Countries.” The aim of the projects is to reduce poverty through support for disadvantaged people; to improve the quality of their lives and to reinforce their own development

capacities; and to strengthen civil society and participatory development promotion. The geographic focus is on ‘backward’ regions in India, in order to reach the poorest population. Even though the Indian and Indonesian EU funded, nationally implemented development policies are vastly different, they all have extremely similar objectives in explicitly empowering women and fostering an environment conducive to gender equality. I will analyze the cases through evaluation reports of the implementing organizations, received through the EU delegation in India and Indonesia.

5.5 Concepts and operationalization

In order to examine the relationship between national implementation of EU development policy with regard to gender equality and women’s empowerment and the effectiveness of these policies on the ground, I will operationalize the concepts that are involved.

EU development policy

The EU’s organizational structure exists out of different supranational institutions, which are the European Commission (EC), the Council of the European Union, the European Council,

(28)

28

the Court of Justice of the European Union, the European Central Bank, the European Court of Auditors and the European Parliament. The EU action on development is based on the Lisbon Treaty and on the European Consensus on Development, which commit the European Commission, the European Council and the European Parliament to a common vision.9 The European Consensus on Development is a policy statement on eradicating poverty and building a fairer and more stable world through reducing poverty, promoting democratic values and implementing nationally-led development by the beneficiary countries

themselves.10_{The EU emphasizes that they are committed to make their development aid} policies more effective through better coordination on the ground, and state that national ownership, donor coordination and harmonization as well as alignment to recipient country systems and results orientation are core principles to achieve effective aid.11_{Additionally, the} EU promises to coordinate their positions in the UN and other international financial

institutions to ensure effective multilateral action and increase development aid effectiveness.

Nationally implemented development policy is thus policy that is implemented at the national or local level, by implementing organizations that are either national organizations, or by implementing organizations that are based in Europe, in partnership with national

organizations. The EC underlines the importance of the ownership of development

cooperation initiatives by the women beneficiaries themselves, through participation in civil society organizations and Community-based organizations, and promotes the emergence of civil society organizations supporting gender equality where they do not exists and ensures capacity-building of the existing civil society organizations.12

Critique from the scholarly community includes the idea that policymakers continue to create policies based on the notion of the 'male breadwinner' as efforts to promote the productivity of the poor are largely targeted to men while women are expected to carry on contributing to household livelihoods and caring for the family with little or no recognition for their efforts and thus means that women’s breadwinning role has been neglected (Kabeer 2003: xiv, 107). This can be seen in the UN initiative based MDGs, where women continue to be perceived

9_{See European Development Policy} (https://ec.europa.eu/europeaid/policies/european-development-policy_en)

10_{See European Consensus on Development} (https://ec.europa.eu/europeaid/policies/european-development-policy/european-consensus-development_en)

11_{See European Consensus on Development C 46/5}

(http://ec.europa.eu/development/body/development_policy_statement/docs/edp_statement_oj_24_02_200 6_en.pdf)

12_{See Strategy for Gender Equality in Development Policy}

(29)

29

primarily in terms of their reproductive roles. There is no explicit mention of gender

inequality in relation to poverty eradication, but gender equality rather appears in relation to health and education in the MDGs (ibid.: xv). Kabeer claims that greater importance should be given to women’s economic contributions in the design of policy because of the

significance of women's contributions to the livelihoods of low-income households, and because focusing on men as breadwinners and improving male earnings does not necessarily lead to an equivalent improvement in the welfare of household members (ibid.: 145). Policy such as the MDGs should therefore focus more on equalizing economic opportunities to improve women's economic agency and earning capacity (ibid.). Scholars like Kabeer, Elson and Cagatay argue that these biases lead to policies that lack explicit recognition of

inequalities of access to market opportunities and are therefore unlikely to promote gender-equitable forms of growth that the MDGs demand from the international community (Kabeer 2013: 213). Kabeer's main critique with regard to the MDGs is thus that enhancing women's access to various resources and ensuring that greater social value is given to their

contributions would make an enormous difference, and that this lies within the domain of policy-makers (Kabeer 2003: 232).

Due to restrictions of time and space, I will explicitly only focus on the national

implementation of policy in relation to effectiveness in Indonesia and India. I will not focus on how development policies are conceptualized by the EU and facilitated towards recipient countries, but assume that those processes can be seen as separate to the national

implementation as this implementation occurs in partnership with national and local organizations. As Schech and Mustafa note, most literature does not focus on how

governments and local NGOs and grassroots organizations conceptualize, design and manage certain development policies as gender inequalities and gender politics are played out in different cultural, political and socio-economic contexts than in comparatively wealthy, democratic Europe (2010: 113). Therefore, it is both relevant and important to look into policies when they are nationally implemented, on the ground.

Gender equality

Gender equality has been defined in different ways in development literature, but the general consensus states that ‘gender’ refers to the rules, norms, customs and practices by which biological differences between males and females are translated into socially constructed differences between men and women and boys and girls (Kabeer 2003: 2). This results in the two genders being valued differently and in their having unequal opportunities and chances

(30)

30

(ibid.). Gender equality then relates to the equality of treatment under the law and equality of opportunity, as well as substantive equality and equality of agency (ibid.). Substantive equality means that the different circumstances and characteristics of men and women lead to unfair gender-related outcomes, and equality of agency means that both women and men are enabled and empowered to make strategic life choices for themselves (ibid.). Because gender inequality cuts across all other forms of inequality, such as class, caste and race, gender inequality exists at all levels of society and makes the effects of poverty worse for women than for men (ibid.).

The UN’s gender inequality index

The UN states that gender equality means that the interests, needs and priorities of both women and men are taken into consideration while the diversity of different groups of women and men are recognized, as well as that gender equality is a human rights principle and a precondition for sustainable, people-centered development (UNESCO 2014). The UN's Gender Inequality Index (GII) measures gender inequalities in three different ways relating to human development. Firstly, the GII measures reproductive health with maternal mortality ratio and adolescent birth rates as indicators. Secondly, the GII measures empowerment with proportion of parliamentary seats occupied by females and proportion of adult females and males aged 25 and older with at least some secondary education as indicators. Thirdly, the GII measures economic status as labor market participation with labor force participation rate of female and male populations aged 15 years and older as indicator. The GII therefore sheds light on the position of women and yields insights in gender gaps in order to create awareness and stimulate critical policy interventions.

Women’s empowerment

Women’s empowerment is seen as an important goal in international development, but the meanings and terminologies of the concept as well as the methods for measuring and tracking changes in level of empowerment are difficult to establish (Malhotra and Schuler 2005: 71). Power is classically seen as a relational concept and defined as the ability to get another actor to do what it would not otherwise have done. However, we can also think about power in terms of the ability to make choices (Kabeer 2005: 13), in which case empowerment is seen as the expansion of freedom of choice and action (Narayan 2002: v). The World Bank therefore defines empowerment as “the expansion of assets and capabilities of poor people to

participate in, negotiate with influence, control, and hold accountable institutions that affect their lives” (ibid.: vi). To be disempowered means to be denied choice, while empowerment

(31)

31

refers to the processes by which those who have been denied the ability to make choices acquire such ability (Kabeer 2005: 13). People who exercise choice in their lives may be very powerful, but they are not empowered as they were never disempowered in the first place (ibid.: 14).

The empowering approach to development is grounded in the idea that poor people

themselves are the most important partners for development, as they are the most motivated to move out of poverty (Narayan 2005: 3). Empowerment approaches can strengthen good governance and promote engagement and agency of citizens (ibid.). However, women's empowerment is often seen as more complex than empowerment an sich, as women are a cross-cutting category of individuals that overlap with various disempowered groups (such as the poor, ethnic minorities etc.) and because interfamilial relations are central to women’s disempowerment in a way that is not true for other disadvantaged groups (Malhotra and Schuler 2005: 71). Whereas empowerment generally requires institutional transformation, women’s empowerment requires specific transformation of patriarchal structures (ibid.: 72). According to Malhotra and Schuler, women’s empowerment as a concept distinguishes itself from gender equality as a concept, as women’s empowerment has two defining features. First of all, women's empowerment explicitly refers to a process of change and towards greater equality, or greater freedom of choice and action (ibid.). The second defining feature is agency, which means that women themselves must be significant actors in the process of change that is being described or measured (ibid.).

Because women’s empowerment is a multifaceted and multidimensional concept that focuses on process and agency, it is also difficult to measure. One of the main problems in measuring women’s empowerment is deciding to which extent there is (positive) change in the process as well as taking context into consideration. Context is crucial to the measurement of women's empowerment as certain indicators and attributes that signify empowerment in one context often have different meanings elsewhere. However, applying indicators across cultures can be useful for making an international or interregional comparison and studies should therefore allow flexibility in the specific indicators used (Malhotra and Schuler 2003: 76, 78). For this thesis, that means that the difference in context of Indonesia and India will be taken into consideration.

Due to its multidimensional character, Malhotra and Schuler argue that measurement of women’s empowerment must extend beyond single indicators or indexes and instead be

(32)

32

analyzed at the micro, meso and macro level to be able to assess the impact of programme and policy efforts (ibid.: 84). The micro level can be assessed by focusing on individual or

household situations, and the meso level can be assessed by focusing on institutional

structures and normative change within the community. The macro level can be assessed by looking at national and international indicators as well as state and civil society institutions. Therefore, we must take into consideration that processes of women’s empowerment can occur through individual behavior, normative change, or collective action (ibid.).

The UN’s MDG 3 to promote gender equality and empower women did use specific targets and indicators to assess its progress. The target was to eliminate gender disparity in primary and secondary education, preferably by 2005, and in all levels of education no later than 2015. The MDGs are often criticized for reducing gender inequalities to single and universally agreed set of priorities and therefore being too narrow in its approach (Kabeer 2005: 23). However, it is also argued that the indicators embodied in MDG 3 have the potential to make a difference as each indicator focuses on immediate change as well as long term consequences and have transformatory potential that have the ability to lead to a genuine expansion of women's choices (ibid.). Kabeer also argues that in order for this to happen, it is crucial that policy changes are implemented in ways that ensure women’s agency - which means that women themselves can participate, monitor and hold relevant actors such as policy makers accountable for their actions (ibid.). For example, if women's access to education has

improved but that education does not provide them with the analytical capacity and courage to question unjust practices, it will not be likely that processes of agency and long term change will occur (ibid.).

Policy effectiveness

Most of the literature on development policy effectiveness focuses on economic development and financial growth and reform, and has generally resulted in unclear and ambiguous results (Bourguignon and Sundberg 2007: 316). Aid skeptics have argued that foreign aid does not promote economic growth because recipient governments misspend the money given by Western donors (Bearce and Tirone 2010: 838). Citing cross-country regression analysis, Bourguignon and Sundberg conclude that the relationship between aid and development outcomes is fragile and often ambiguous, which is explained by the fact that much aid is lost due to instability and conflict as well as that much aid is often not for developmental purposes but for disaster relief or military or political ends (ibid.). Based on this knowledge, new ideas and changes in aid architecture are emerging. For example, the emphasis on country

Implementation of EU development policy on the ground : a comparative study on India and Indonesia