“Who decides how money gets spent in your household?”
Researchers focused on measuring women’s economic empowerment (WEE) have often asked this question (paraphrased above) through household surveys to gauge women’s level of agency and decision-making power relative to their spouses and other family members.
But what if we have been asking the wrong question?
Decision-making over household expenditures may be the wrong measure of WEE in some contexts. For women in poor households in Latin America, making everyday decisions about household consumption is part of their traditional role as caretakers—not a sign of empowerment. Women informal workers who belong to the Self-Employed Women’s Association (SEWA) in India concur: decision-making over household consumption is a foreign concept and not a good fit. For them, full employment—that which provides a secure income, labor protections and quality benefits—is a better measure of economic empowerment.
Decision-making over household expenditures can also be subject to self-reporting biases. Economic empowerment has both an objective dimension expressed in changes in employment, income, and assets (among others), and a subjective dimension. The subjective side of empowerment (the expression of agency) is mostly unobservable and depends largely on self-reporting, which can be unreliable.
Does this mean that it is close to impossible to identify a reliable measure of WEE that can be widely used across contexts and cultures?
A sense of the possible emerged at a meeting that CGD and Data2X held at the end of April—the fourth such meeting in a collaboration that began in 2015, when we discussed the UN Foundation and ExxonMobil Foundation report Measuring Women’s Economic Empowerment. We invited researchers and practitioners, the latter from both government and the private sector, with the aim of strengthening the linkages between academics who are focused on conceptualizing, building and testing WEE measures and policymakers and private sector actors who want to use rigorous WEE measures to understand the impact of their programs and investments.
Defining and measuring WEE
The last 5 years have seen a commendable surge in the number of research initiatives seeking to define and measure the different dimensions of WEE. Notable research efforts include:
The Women’s Empowerment in Agriculture Index (WEAI), a pioneering index first launched in 2012. The index, its most recent version being used in 54 countries, measures 3 domains of women’s agency: intrinsic agency (power within), collective agency (power with), and instrumental agency (power to). Researchers behind WEAI are now working on a new WEE metric for use by national statistical systems.
- Women’s Empowerment: Data for Gender Equality (WEDGE), an initiative launched in 2019, seeks to integrate theoretical concepts from feminist academic literature with WEE measurement through experimentation and the development of scalable models. Its final objective is to increase dialogue between experts and policymakers.
- The Abdul Latif Jameel Poverty Action Lab (J-PAL) has a number of work streams on WEE measures, including a useful methodological guide and a special fund for measuring innovations for use in randomized control trials (RCTs). Researchers leading a new study in India are developing a short (5 question) module to measure women’s agency, validating questions from the ground up.
- The Growth and Economic Opportunities for Women (GrOW) program of Canada’s International Development Research Centre (IDRC), in its second round, attempts to embed context-specific measures into global frameworks. Their recent literature review demonstrates the importance of retaining nuance while developing locally relevant measures.
- Evidence-based Measures of Empowerment for Research on Gender Equality (EMERGE), a database of broader women’s empowerment measures, includes a focus on WEE among other areas, as well as information on the reliability and validity of different indicators related to women’s employment, asset ownership, etc. EMERGE researchers have created a conceptual framework to measure WEE, focusing on measuring social norms and agency, and seeks to validate these measures to use at scale.
- The World Bank’s Africa Gender Innovation Lab (Africa GIL) has developed context-specific measures of WEE (“going deep”) and a set of standardized indicators (“going broad”), and looked ahead into cutting-edge measures in partnership with other researchers (“going forward”).
Promoting WEE through policy and practice
As researchers attempt to define and measure WEE, more policymakers and private sector actors have turned their attention to promoting WEE through their projects and programs. These include:
- Women Entrepreneurs Finance Initiative (We-Fi), a World Bank-housed initiative dedicated to supporting women-owned small- and medium-sized enterprises through access to finance, capacity building, and markets, as well as engagement with policymakers to remove legal and regulatory barriers women entrepreneurs face;
- Women’s Global Development and Prosperity Initiative (W-GDP), a United States inter-agency initiative aimed at supporting women’s workforce development and entrepreneurship, as well as establishing enabling environments to promote WEE;
- Walmart’s Shared Value Approach seeks to identify the barriers that women face in global value chains and leverage its purchasing power to act as a lever of change;
- The United Kingdom’s Department for International Development (DFID) is working to identify a sub-set of WEE indicators to improve the way it records and measures WEE globally. DFID conducted an internal review of WEE related indicators in 119 economic development programs and is currently working on a guidance note to measure WEE.
On the whole, we see a positive trend towards convergence in WEE research and measurement efforts. This is partly resulting from intentionally built research partnerships —such as a partnership between the World Bank AGIL with IFPRI (WEAI) and the University of Oxford. And through meetings like the CGD and Data2X held, we hope to continue to reinforce and facilitate these connections across different WEE measurement initiatives.
Bridging the gap
But there is disconnect between existing WEE measurement frameworks and what practitioners need.
Representatives from government agencies and the private sector emphasize the need for a concise set of practical WEE metrics, and those that can be easily shared with and used by colleagues who are not gender specialists. This producer/user discrepancy is not only the province of WEE measures. It is a more general data issue that requires establishing a more systematic purposeful dialogue between researchers and practitioners and between researchers and national statistical offices (NSOs). WEAI’s current work stream of creating a new WEE metric specifically to be adopted by NSOs, for example, is a move in the right direction, as is the Africa Gender Innovation Lab’s work to “go broad” by developing a set of standardized indicators that will allow for comparisons across contexts.
Two additional challenges are measuring the objective dimension of WEE and measuring social norms shaping WEE. While much of researchers’ efforts have gone towards defining and measuring the complex concept of agency, there are still issues with measuring “objective” dimensions of WEE, including women’s labor force participation, partly as result of gender bias in the way questions are defined and information is collected.
WEDGE found substantial undercounting of women’s paid work in a pilot in India when questions were phrased in terms of principal occupation versus specific activities. An ongoing collaboration between Data2X, ILO, and the World Bank on measuring work and employment in Sri Lanka shows similar results: measures of women’s paid work are very sensitive to the way the questions are phrased. Even a slight change in questions’ phrasing can increase the counting of women’s work.
Measuring social norms affecting WEE is another challenge. Social norms are used extensively to explain things away, so much so that the concept is becoming a “catch all” and losing explanatory power. A first step is to break down the complex concept into specific attitudinal and behavioral dimensions related to WEE and operationalize and measure them separately. The work of the World Bank Africa GIL on women “crossing over” into traditionally male occupations is a good example of this translation of a social norm into its attitudinal and behavioral referents.
Going forward, researchers should prioritize a coordinated approach, informed by the needs of practitioners. Despite the complexity of WEE, including the conceptual frameworks that are sometimes built around it, most measurement efforts are intent on finding streamlined solutions:
Of the organizations we’ve highlighted above, WEDGE researchers are seeking to develop models that can be scaled; GrOW is looking to embed context-specific measures within global frameworks; EMERGE has identified 16 higher scoring measures of WEE agency; J-PAL is creating a short module to measure women’s agency; the World Bank Africa GIL has a set of about a dozen standardized indicators; and WEAI has simplified its metrics in newer versions of the index.
Building on these impressive efforts, researchers should come together to agree on the attributes of suitable WEE measures and test them for validation. If a global WEE measure is not possible because of differences across cultures and contexts, a few measures may be possible across regions or major categories of work (like agriculture, self-employment and entrepreneurship, or wage and salaried employment).
This coordinated approach needs to include working with UNSD and ILO Statistical teams who work with NSOs on international standards, as well as governments and the private sector as users of these measures. Donors and international agencies, especially those with robust gender data programs, such as the World Bank and UN Women, should participate in and support this coordinated effort.
A set of reliable, comparable WEE measures is foundational to improving the evidence base on what works to narrow gender gaps in the economy and promote women’s well-being. It is also essential to the success of policies, government programs, and private sector investments that aim to do the same. With COVID-19 top of mind, understanding and responding to women’s economic vulnerabilities and contributions is essential, and the development and use of solid, practical WEE measures is central to this objective. The benefits of having good WEE metrics will last.