Rationale for the Gender Equality Index for Europe

Several important international gender equality indexes have been identified and reviewed by the study: the Gender Inequality Index of the UNDP, the Gender Gap Index of the World Economic Forum, the Gender Equity Index of Social Watch and the European Union Gender Equality Index (EUGE-Index). A comparison of these indexes shows that there appear to be some unanimity on the methodological and conceptual level. Three of the four indexes try to measure equality as such, by concentrating on the absence of gender gaps. In addition, there is a large overlap in the choice of relevant dimensions (perhaps partly inspired by the availability of data). Economic participation is clearly important, as well as education /knowledge, money and power. Health is included in two of the four indexes; (equal division of) time in one. Of course, the differences in approach and design have an impact on the outcomes. Yet, the scores of the indexes show large similarities in the ranking of the countries.

A number of conceptual issues are discussed as part of the study. A fundamental requisite in the development of the index is that it must be based on a sound conceptual framework and not only data requirements. Gender equality is, the study argues, a complicated term with diverse dimensions and many layers of meaning. It may refer to a situation in which women become more equal to men (for example, in terms of labour force participation), or to a more equal valuation of the different gender roles (for example, by taking into account the time women spend on care). In addition, it may refer to a situation in which gender roles and structures are seriously challenged (by changing the current gender specific organisation of labour and care). Gender equality is seen as implying a change in the lives of both women and men through the promotion of greater equality in the distribution of paid and unpaid work. The broad perspective also infers that an equal distribution of paid and unpaid work is not sufficient. A full concept of gender equality should also take into account the political dimension and physical integrity of both women and men.

An important issue raised by the study is the definition of gender equality. In some documents there is a heavy reliance on notions of sameness, whereas in others, the emphasis is more on the equal valuation of differences and/or the transformation of existing gender practices. In general, most researchers seem to agree that gender equality has different dimensions which cannot easily be summarised in one single uni-dimensional indicator. In the Beijing Platform for Action, for example, 12 critical areas of action were listed, ranking from ‘women and the economy’ and ‘violence against women’ to the ‘institutional mechanisms for the advancement of women’ and ‘women in the media’.

This rather extensive perspective on gender equality as a particular combination of sameness, difference and transformation, is also reflected in the Women’s Charter (EC 2010a). It presents a series of commitments of the European Commission, based on agreed principles of equality between women and men:

• equality in the labour market and equal economic independence for women and men, namely through the Europe 2020 strategy;
• equal pay for equal work and work of equal value by working with Member States to significantly reduce the gender pay gap over the next five years;
• equality in decision-making through EU incentive measures;
• dignity, integrity and an end to gender-based violence through a comprehensive policy framework;
• gender equality beyond the EU by pursuing the issue in external relations and with international organisations.

These five priorities are translated into concrete measures in the Commission’s Strategy for Equality between Women and Men 2010-2015. The measures range from getting more women into company boardrooms and tackling gender-based violence to raising the transparency of pay structures. The Strategy underlines that gender equality is not equated with equal positions of women and men within paid employment (equality-as-sameness strategy), but embedded into a more broadly defined ‘gender equal regime’, as a result of which relevant policies might also refer to, for example, equal sharing of power between women and men, reducing the school drop-out rate for boys, promoting fathers access to parental leave and combating domestic violence.

Another issue important raised by the study is whether the index is supposed to measure equality as such or rather a mixture of welfare, development and equality. Quite often for the sake of clarity the focus is on gender equality as such, being defined as the absence of gender gaps on certain domains. This decision has rather large consequences, though. Several important gender equality domains are not easily described as a distributional issue, as a result of which gender differences are not easily described in gaps. Gender-based violence, for example, is clearly not a distributional issue.

These issues provide a real challenge to all attempts of measuring gender equality. Not only because of the difficulty of conceptualising equality, but also because of more practical issues such as finding the right indicators and sex-disaggregated statistics.

The study reiterates that the design of an index is always a compromise between theoretical ambitions and rigour on the one hand and practical considerations on the other. In the case of the Gender Equality Index, the theoretical ambitions refer to the definition of gender equality; the practical considerations refer to the choice of indicators and the availability of sex-disaggregated statistics. 

Choice of Domains

The study points out that the choice of domains is essential. Current indexes in the field of gender equality are based on a rather small set of domains, partly because of theoretical and partly because of practical considerations. From a theoretical point of view it seems important to limit the number of indicators; a large collection of indicators could obscure the most salient developments. Some domains may also be highly correlated, for example educational attainment, paid labour and risk of poverty, which raises the question of the added value of having all variables included. In this respect it is also important to differentiate between dependent and independent variables. Childcare facilities, for example, or flexible working hours can be seen as important provisions in order to promote women’s full-time participation in the labour market. However, they should not be treated at the same level as outcome variables such as labour market participation or decision-making power. If they are, there is a real danger of double counting and overestimating the differences between countries.

Another sensitive choice highlighted by the study is based on what the index can be designed to measure: either the current state of (in)equality in a country or the factors that contribute to its achievement. An index oriented to measuring the existing state or level of gender (in)equality in a country cannot combine outcome variables with instrumental variables: the risk is double counting or in under/overestimating the phenomenon or, at worst, violating the conceptual coherence of the framework. 

Weighting

Another important methodological issue refers to the weighting of domains. In most indexes, the indicators and domains get equal weight when calculating the composite score. However, due to differences in spread of values, indicators may implicitly have a different weighting: the higher the standard deviation an indicator has, the more weight this indicator gets in the final composite score.

Gender gaps

For the construction of the Gender Equality Index, the study makes recommendations for dealing with gender gaps. One type of gap is the result of a situation where men outperform women; for example in the field of paid employment in most instances the scores of women are below the scores of men. Another type of gap implies that women outperform men. This is quite often the case in the field of education, especially among the younger generations where the educational attainment of women is higher than that of men. A focus on gender equality, which is operationalised as the absence of gender gaps, would imply that both types of gaps are to be treated the same. Not doing this might result in a high ranking of countries in which men are in a disadvantaged position.

Monitoring in time

The study argues that an important requisite of the Gender Equality Index is that it should enable the monitoring of progress over time. An increase of the composite index score would indicate a narrowing of the gender gaps, whereas a decrease would imply increasing gender gaps. Overall, monitoring gender equality over time appears to be complicated. The possibilities of monitoring are limited due to the lack of harmonised data. As a result, monitoring is only possible based on a few domains, which unfortunately gives only a partial picture of gender equality.

The potential of a slightly different organisation of the Gender Equality Index can be explored, according to the study, by developing the idea of a core index with satellite accounts. Setting up a satellite account may be extremely useful in order to explore the state of the art on a domain that is not easily integrated into the core index, either because of lack of data, for example in the case of exploring the domain of the quality of work, or because this concerns a more instrumental domain which is not easily combined with the outcome related indicators of the core index such as child care facilities, or even because this domain is not easily interpreted as a distributional issue which might be the case of gender-based violence. As such, the study recommends setting up a satellite which can be interpreted as an important reporting and mutual learning system, as a stimulus to collect data in a cross national comparative way and as a policy tool to analyse what EU Member States do in important domains of gender equality.

Given the difficulties of providing a full measurement of gender equality, the study suggests to organise the gender equality index in a more flexible way by defining a core index and some satellite accounts. The core index could concentrate on a small number of equality domains and could be used to monitor development in time on a regular basis. In addition the satellite accounts could be used to analyse other important issues on some domains which are not easily defined in distributional terms and refer to instrumental variables.