× My Blogs Create New Blog Change Password

Assessing Gender Inequality through Sen's Capabilities Approach- A research proposal

By Megha Bhattacharyya ⊚


The traditional approach to measuring and gender inequality involves examining market outcomes such as gender-based wage differential. However, in a society with patriarchy as institutionalized as India’s, newer approaches must be taken to assess the same. Sen’s elaborations (Klitgaard & Sen, 1985; A. Sen, 1995; A. K. Sen, 1994; Sugden & Sen, 1986) on his capabilities approach has immense potential to provide insight into the problem of gender inequality, and can be used to demonstrate methods to enable female empowerment, which would result in the restoration of social justice. Sen’s capabilities approach while largely used for examining poverty, can also be extended to measuring inequality, since the approach essentially employs methods to examine deprivation of many kinds. 

This paper will largely attempt to observe the extent and impact of gender inequality through Sen’s capabilities approach in India.






Sen’s capabilities approach is a framework to analyze, assess, and evaluate poverty, inequality, development and related issues, thereby focusing on what people are effectively able to do and be, instead of just their disposable income, consumption or expenditures (Klitgaard & Sen, 1985; Robeyns, 2006a; A. Sen, 1995; Sugden & Sen, 1986). The core idea of the approach involves capabilities and functionings. Functionings are defined as the beings and doings of a person, while capabilities are defined as the ability to achieve these functionings to achieve the kind of standard of living they aspire for. Capabilities are the potential, while functionings are the realized. Through this approach, therefore, welfare is viewed as a multidimensional issue, quite pluralistic in nature. This implies that the examination of the same should be treated as such. The focus shifts from simply looking at income and consumption data to a vast array of indicators that portray the different dimensions of well-being. Such indicators are constitutive elements of people's lives, such as knowledge and education, health, nutrition, etc. which should not be ignored while assessing inequality (Martinetti, 2000). Sen argues that exclusively focusing on the income aspects of inequality is a gross oversight, in that they are a crude measurement of well-being, and that shifting attention to one’s ability to freely live a valuable life and enjoy positive states of being are a better reflection of the same.

Therefore, when such normative evaluations are made, we must take into account what an individual can do or be, rather than their consumption and disposable income. Two individuals could very well be at the same level of income, but not be able to realize (achieve the functioning) the potential (capabilities emanating from the use of the income). Monetary measures of inequality often overlook the nuances of societal, personal or institutional differences which result in the difference in the standard of living of two individuals with the same wealth status. Discrimination based on gender, race, class, and religion is often resultant in limiting the opportunities available to certain groups of individuals. The impact of such discrimination against women on the opportunities available to them, and whether such opportunities are indeed enough to provide them with the freedom to live a valuable life or the kind of life they prefer, is indeed an aspect that needs to examined thoroughly.




The empirical studies conducted by Amartya Sen, which essentially laid the foundation for the approach were validated by leading commentators in the field in the years to follow (Kuklys, 2004; Kuklys & Robeyns, 2005; Robeyns, 2005). He concluded from his empirical observations of the GNP and overall standard of living in Brazil, Mexico, India, China, and Sri Lanka, that even though the GNP per capita was highest in the two, the latter three fared better in terms of life expectancy at birth, infant mortality and child morbidity, Sri Lanka in particular. This subsequently led him to infer that policies in Sri Lanka were designed to attain efficient distribution.

Sen’s approach thereafter has led to a marked improvement over the conventional manners of measure of well-being. Several socio-economic indicators to measure inequality, including the Human Development Index, and its successor Human Poverty Index, have been derivative of the capabilities approach. As suggested by the approach itself, such indices essentially measured the level of deprivation suffered by certain countries, and certain groups within those countries. Klasen (2000)and Qizilbash & Clark (2005) extended the capabilities approach to South Africa and used fuzzy sets poverty to examine deprivation at the household level. Martinetti (2000) used the fuzzy poverty theory to examine deprivation in Italy. She concluded from the results of the survey that while inequality in the functionings space existed, it was much less than that in the income space. This conclusion renders income inequality to be quite a grey measure for deprivation, in that it is crude and could potentially understate or overstate the actual level of deprivation.

The measurement of women’s agency is essential for gender equality not only in the market-based transactions but also in the status of a woman within her family and society. Amartya Sen set off the metaphorical wheel in motion for the examination of gender inequality through the capabilities approach in his examination of sex bias in India (A. Sen & Sengupta, 1983). He found that females have worse achievements than makes for functionings such as age-specific mortality rates, malnutrition, and morbidity (Klitgaard & Sen, 1985; A. Sen & Sengupta, 1983). More recently, leading commentators have begun using Sen’s capabilities approach to measure, assess and analyze aspects of gender inequality either in part or as a whole (Agarwal, Humphries, & Robeyns, 2003; Alkire, 2002; Robeyns, 2003, 2006a). Evidently, the capabilities approach is of enormous importance to several feminist concerns. 

‘It advocates empowerment and draws attention to the critical role of social, political, legal and economic institutions in reducing the deprivations and enhancing capabilities over time’ (Alkire, 2008). Sen has claimed ‘that the question of gender inequality…can be understood much better by comparing those things that intrinsically matter (such as functionings and capabilities) rather than just the means to achieve them, like resources. Gender inequality is just an issue of disparate freedoms’(A. K. Sen, 1994). Largely since its inception, the idea of reduction of deprivation has been central to the feminist movement. More importantly, most issues pertaining to the movement, such as domestic violence, reproductive healthcare, education, voting rights, etc. cannot be reduced to monetary welfare, even though financial independence and equal pay is a part of the movement itself (Robeyns, 2003).


The advantages, therefore, of using the capabilities approach to examine the overarching feminist concern, gender inequality, are many. Firstly, from an ethics viewpoint, the capabilities approach is individualistic in nature, thus rejecting the idea that the well-being of an individual can be subsumed by unit larger than themselves (such as a household). The well-being of each person is considered separately and it lends greater importance to personal characteristics, social relationships and an inter-disciplinary interaction between the various aspects of well-being. It, however, does not discount care and interdependence, and this is a desirable characteristic and inequality analysis. Secondly, the capabilities approach is not restricted to market outcomes, in that it not only takes into account the well-being within the markets but also outside the markets. In the context of gender inequality, unpaid domestic work and non-market care of others, the onus of which largely falls on the shoulders of the women, is also take into account. In a purely market-oriented assessment, an integral part of the "double burden" that most women have to bear, is largely left out. Thirdly, the diversity in race, gender, age, physical and mental characteristics are also accounted for, thus making it easier to understand why the deprived are where they are (Robeyns, 2006a). 




In light of the current literature, there are some lacunae that need to be addressed. There is a general lack of consensus with regards to the capabilities and domains to be considered, and whether or not there is a need for one single, definite list. The methods for measurement of decided capabilities also vary from one piece of literature to another. Sen has not defined any particular set of capabilities to be followed and only provides a general framework for examination. One could argue that following the capabilities and capability-like dimensions used by Humphries (1993) and the likes could pave a definitive path to analysis. However, since our aim is to examine individual deprivation, indices developed at a macro-level are not necessarily consequential to us (Robeyns, 2003). The question of whether or not a single canonical list is required however still remains. While a one-size fits all list of dimensions and capabilities seems attractive in that it maintains uniformity and potentially allows for faster, less laborious analysis, several considerations must be made before one can draw a conclusion in this regard. Nussbaum (2000) defends that a list of central capabilities maintains a critical edge in the research conducted. She argues that in order to be able to apply the capabilities approach to matters of social justice and gender inequality, a definite list must be endorsed (Nussbaum, 1998). She defended the need for a singular list, maintaining that such an arrangement would provide literature with the critical force that is required. Nussbaum (2000) and Glover & Nussbaum (2003) have identified and defined a list of such capabilities that she claims are universally valid and are central human capabilities. Nussbaum concedes to the idea of adaptation by context, but she continues to firmly maintain that the specification of a universal list of capabilities should essentially be the first step. Her point of defense is the idea that if left too open-ended, the possibility of wrong freedoms being prioritized becomes higher. Nussbaum's approach, therefore, involves a universal general list, which can be modified according to context. 

Alkire (2008), Robeyns (2003) and A. Sen (2004) disagree and argue against the construction of a canonical list that needs to be followed. Alkire (2002) notes that Nussbaum's list is constructed at a very abstract level. It is a prescriptive list of things-to-do and is a list essentially used to create a universal theory of good, and appeals to the world as a whole. Sen's approach, on the other hand, is under-specified and has a broader scope. Sen's argument is that lists of capabilities should be context-dependent and the definition of the same must be done as such. Since the capabilities approach can be used for a variety of evaluative exercises and purposes, it is imperative that lists remain context-based and that there cannot exist a single, canonical list. A fixed list is inappropriate in this regard. It must depend on the geographical area in which the evaluative exercise is being conducted, the purpose of the exercise and the nature of the evaluation that follows, and therefore must be tailored accordingly. Appropriate elements of the lists will also largely depend on prevailing social conditions and the people's understanding and engagement in social issues plaguing their lives (Alkire, 2008).

Evidently, all three of the leading commentators in the field concede to the idea of further specification in these lists before use. Sen’s retort to an explicit set of capabilities is “I have nothing against the listing of capabilities, but must stand up against a grand mausoleum to one fixed and final set of capabilities” (Alkire, 2008; A. Sen, 2004a). There is a need to select dimensions and capabilities before the capabilities approach can be applied to any matters of social justice. However, the insistence of a fixed list of capabilities would be akin to completely discount the ‘possibility of progress in social understanding and go against the productive role of public discussion, agitation, and protests' (A. Sen, 2004b). A fixed list undermines the reach of democracy, in that it undermines the ongoing public conversation and reason, disregards the idea that societies progress, and ignores the very individualistic nature of the approach. However, one must not have to swing between the extremes of Sen’s complete reliance on public reasoning and discourse and Nussbaum’s definition of a fixed, universal list. The former largely provides no guidance as to how one must go about selecting relevant capabilities, while the latter seems too restrictive to be applied to most matters of social justice.


The beauty of the capabilities approach although, lies in the allowance of the employment of plural techniques, in that one could choose from a plethora of methodologies and analytical techniques, which do not compete with each other (Alkire, 2008). Studies that compare welfare rankings at the macroeconomic level (Balestrino & Sciclone, 2001) find that monetary evaluations of inequality tend to widely differ from the results of their approach. Other studies use individual household-level data, often from household surveys. Notable ones include Brandolini & D’Alessio (1998), Klasen (2000), Kuklys (2004), Martinetti (2000). To measure individual-level inequality, three major types of techniques have been used: fuzzy sets poverty, descriptive statistics, and factor analysis.

Brandolini & D’Alessio (1998) measured well-being in the functionings space at the household level in Italy through the use of descriptive statistics. Since their study, as well as other notable studies of the kind largely involved examining poverty and inequality, descriptive statistics was best suited for them. Their approach was reliant on the evaluation of very few indicators and since their analysis could be done by merely looking at shares of deprivation., the use of descriptive statistics was advantageous in this regard. The main disadvantage of descriptive statistics arises when a large number of indicators are being put to use. Observations become unmanageable and summarization of inferences tends to become difficult. 

Kuklys (2004) made use of factor analysis, which reduces indicators to a small number of underlying variables (Robeyns, 2006a). Factor analysis makes their assumption that indicators are linear combinations of functionings and therefore makes it possible to reduce huge amounts of data to a smaller more manageable number of variables.

Martinetti (2000) uses fuzzy sets theory, a generalized version of scaling to reduce data to smaller amounts. Fuzzy sets theory rescales the value of indicator from 0 to 1, the lowest possible value being rescaled to 0 and the highest possible value being rescaled to 1, while the values in the middle use different functions. Most importantly, unlike standard rescaling techniques, fuzzy sets allow for both linear and non-linear rescaling. Rescaling is dependent on the frequency of deprivation, i.e. a person will have a higher deprivation score if lesser people in that sample are deprived of the same.  


Both factor analysis and fuzzy sets theory are relatively sophisticated techniques of measurement Robeyns (2006b). However fuzzy sets theory makes normative assumptions while deciding the frequency of deprivation. More importantly, both these measures estimate an individual’s functionings score, and subsequently, use multivariate regression analysis to analyze the indicators that have an impact on the functionings index. However, matters of social justice as important as gender inequality cannot and by extension, should not be restricted to mere statistical significance, as it limits the notion of gender. The lacuna observed here is also something that this paper attempts to resolve. 

The question of which functionings and capabilities to take into account still remains. Any measurement of the empowerment of women must include an individual's sense of self-confidence and esteem. This is imperative to not only determine the respect women deserve in terms of market outcomes but also at home, in the society and their families. The following aspects must be addressed:

  1. personal laws that determine the rights of women within the family- especially those related to marriage, divorce, inheritance, etc.

  2. loss of dignity through marital violence and societal slander.

  3. questioning of the patriarchal ideology to justify insubordination. 


By examining the existing literature in the use of capabilities approach to measure gender inequality, a severe dearth of reliable literature in the Indian context was found to exist. Moreover, most empirical studies that have used the capabilities approach to measure inequality are largely focused on poverty as a whole, which leaves us with a much smaller pool of literature in this regard, to refer to in the first place. 

Robeyns (2005, 2006b) has probed into the practical applications of the capabilities approach, among which one was the use of the same in examining gender inequality. In a precursor to her own work in 2006, Robeyns has conceptualized a list of capabilities for application in the developed Western countries (Robeyns, 2003)

To substantiate the theoretical survey, Robeyns (2006a) has used BHPS data to test the gender inequality inherent in British society. The data suggests that women are significantly disadvantaged in the dimensions of physical and mental health, leisure, while men did worse on social relevance. No evidence of inequality was found in shelter, mobility, and education. A very obvious drawback of this survey was that the BHPS doesn't include time autonomy, which is an integral part of assessing the deprivation (whether intentional or not) of women in autonomy.



Although Robeyns (2003, 2005, 2006b, 2006a) has contributed significantly to the conceptualization of a somewhat structured list and has provided a direction for selection of capabilities, it still seems inexhaustive. Moreover, since there is an obvious lacuna of such literature and empirical studies in the study of gender inequality through the capabilities approach in the Indian context, there is a need to fill this void by conducting studies that they employ such an approach.

Robeyns (2003, 2006a) are also restricted largely to post-industrialized western societies, the structure of which is vastly different from that of India’s. While there are certain aspects of certain capabilities that women, as well as men,  can take for granted in the West, it may not be so here. Despite rapid globalization and integration with the world, it remains a fact that certain societal constructs in India hinder the achievements of women in the functioning space. India is the most unsafe country for women in the world. Therefore, while examining capabilities such as shelter and bodily integrity a lot more focus must be placed on the safety of women, and crimes and violence against them (Panda & Agarwal, 2005). It is also fairly evident that the Indian society has a deep-rooted masculine bias, in that male persons are favored over female persons(A. Sen & Sengupta, 1983). Therefore, there is an inherent and widespread inequity in access and autonomy over some of the most basic aspects of life, starting from education to healthcare and access to good reproductive care. It is this masculine bias that also puts numerous women in India at risk of suffering violence, both in and outside their homes. Moreover, women in India are significantly disadvantaged in terms of financial autonomy, in various forms including but not limited to ownership of assets(Doss et al., 2019). Therefore, special attention to such nuances of capabilities must be taken into consideration while testing the deprivation in functionings achievements of women in India. 




The primary objective of this paper is to examine gender inequality in India through the capabilities approach. The justification of adopting this approach is that it encompasses non-monetary aspects of deprivation that women face due to discrimination and bias.

The present paper, therefore, proposes to fill the aforementioned gaps in literature at least in part, if not as a whole by attempting to answer the following research questions by means of an empirical study:

  1. In which capabilities are deprivation for women more significant in?

  2. What is the extent of inequality in the achievement of the decided functionings and capabilities? 




The paper proposes to carry this forward the classification conceptualized by Robeyns (2003, 2006a) in order to simplify the summarization of the findings. The present paper proposes to use questionnaires from the NFHS surveys as references for certain capabilities such as health, autonomy, paid and unpaid work, education, etc. The proposed methodology would be to create an inequality index based on the Alkire-Foster method. In order to get a finer look at the gender inequality prevalent in India, one must tailor a list of capabilities to suit the context. The list of capabilities conceptualized in Robeyns (2003, 2006a) can be considered as a sufficient outline for constructing a list. However, changes need to be made in the nature and line of inquiry when performing empirical studies, in order to not overlook the nuances of Indian society. Proposed data is to be taken from the National Family Health Survey, or the District Level Household and Facility Survey, while macro-level data can be looked for on CMIE and Indiastat.