Guidance Note for Task Teams 76253 Integrating Gender into Poverty and Social Impact Analysis Social Development Department & Gender and Development Group The World Bank February 2013 Poverty and Social Impact Analysis I ntroduction Poverty and Social Impact Analysis (PSIA) is an analytical approach used to assess the distributional and social impacts of policy reforms on dif- ferent groups. PSIA can be undertaken pre- or post-policy reform. If done before or during the reform process, the analysis can provide a sound empirical basis to inform the design and sequencing of alternative policy options. If undertaken after the reform, PSIA can help assess the actual impacts of the policy, which can suggest ways to miti- gate any adverse effects and help decision makers understand the likely impacts of future reforms. PSIA can ensure that decision makers have a strong analytical and evi- dence-based foundation as they make a policy choice, especially if conducted before or during the reform process (see box 1). In addition, it can help create space for policy dialogue around reforms, contributing to increased transparency in policy formulation. Gender is a relevant dimension of policy reform impacts that is largely neglected in PSIA. Reforms often impact men and women differently because they encounter very different opportunities and constraints, and they have different roles and needs. These are influenced, to a great extent, by cultural contexts and deeply embedded social norms, which often work to the disadvantage of women. Land policy and agrari- an reforms, for example, often take place in the contexts of traditional ownership rights and customary division of labor between men and women. The impacts of labor mar- ket and trade liberalization reforms partly depend on the occupational segregation between men and women or the gender gap in earnings. Further, social norms and other constraints in many countries may prevent women from building endowments and exercising agency, which can limit their participation in decisions about policy re- forms that affect their lives.¹ Gender integration in PSIA refers to the process of understanding the implica- tions of a policy change or a reform for different groups of women and men, as well as the dynamic relationships among them. The underlying premise is that dif- ferent groups of men and women have different needs and roles in society; that each group is affected differently by social, political, and economic processes; and that poli- Box 1. PSIA as “upstream� input into policy design The PSIA on “Distributional Impact of Electricity Tariff Hikes� in Bangladesh and Nepal analyzes how tariff struc- tures impact political and social outcomes. The findings influenced the discussions in the Financial Recovery Plan for Electric Utility in Nepal.* A PSIA in Thailand on tax incidence similarly informed the World Bank’s tax policy review and advisory report to the government at a time when the Thai Ministry of Finance was develop- ing a comprehensive tax reform. Prior to the PSIA, the assumption had been that the Thai value-added tax (VAT) was a regressive tax and should not be raised. But the analysis suggested that the VAT in Thailand was indeed slightly progressive because the poor consumed VAT-exempt goods and mostly bought goods and services at nonVAT-registered traders. So, following the PSIA, the Ministry of Finance proposed raising the VAT. Similarly, the PSIA showed that personal income taxes were progressive—which has fed into other recommendations made by the Ministry of Finance.‡ * See World Bank, 2012, “Poverty and Social Impact Analysis Multi-Donor Trust Fund Annual Report, July 2011–June 2012� (Washington, DC: World Bank), 6. ** Ibid., 22. 2 Guidance Note cy interventions may need to be focused and tailored in order to address these differ- ences. PSIA that recognizes the gender dimensions of reforms—and their distinct wel- fare and social impacts on different groups of men and women—can inform policy interventions, so that they can take these gender differences into account. This, in turn, has the potential to improve policy effectiveness and impact. Finally, a gender perspec- tive in policy design and implementation can help identify opportunities for redressing gender constraints, therefore promoting greater gender equality, which is a fundamen- tal development goal and human right. Given the World Bank’s renewed corporate commitments to gender, this note aims to facilitate more gender-informed analysis through PSIA in Development Policy Operations (DPOs). In doing so, this note complements the Good Practice Note on Integrating Gender into Development Policy Loans. This good practice note points to underlying analysis as a major opportunity for ensuring that DPOs are gender- informed in terms of design, monitoring, and evaluation. Box 2 provides some exam- ples of gender analysis that informed policy formulation. PSIA is considered one of the main analytical tools for informing the design of policy reforms, and DPOs in particu- lar.² Yet, since current PSIA are often not gender-informed, they can rarely provide an Box 2. Gender analysis increasingly informs policy interventions A country social analysis in Nepal* examined how gender, caste, and ethnicity influenced individual and group political power and economic opportunities. The analysis documented caste-, ethnic-, and gender-based dispari- ties found in different dimensions of human development, including education levels, life expectancy, child mor- tality, access to services, and political representation. The policy recommendations contributed to a number of national policies and laws targeting gender exclusion in particular. The Interim Constitution of 2007 guarantees social justice and affirmative action for women. Physical or mental violence towards women is now punishable by law and there is a law against marital rape. Another initiative triggered by the analysis was the improvement of the monitoring and information systems of the Ministry of Education and the Ministry of Health and Popula- tion, so that they can better collect and analyze data for different social groups. The country gender assessment for Chile** focused on the issue of low female labor-force participation, its determinants, and potential economic costs. The analysis identified the lack of specific and continuous experi- ence as one of the determinants for why women in low-income households stayed outside the labor force. The economic and sector work was used to design a component in a social protection operation to finance skills development (Chile Califica) that specifically targeted women who were out of the labor force in remote poor areas in Chile. A PSIA in the Democratic Republic of Congo (DRC) examined an intervention to ameliorate border conditions in the eastern part of the country. As part of this exercise, the PSIA analyzed the impact of training on the em- powerment of female cross-border traders to stop harassment at the DRC’s border with Rwanda. A video that was produced as part of the PSIA supported discussions on trade facilitation at the African Union Summit in January 2012, which led to the adoption of the “Traders Charter.�‡ * L. Bennett, 2006, “Unequal Citizens: Gender, Caste, and Ethnic Exclusion in Nepal� (Washington, DC: World Bank). The text for this box is taken from the online Country Social Analysis Toolkit (currently only available to World Bank staff). ** World Bank, 2007, “Expanding Women’s Work Choices to Enhance Chile’s Economic Potential: Chile Country Gender Assess- ment� (Washington, DC: World Bank). ‡ See World Bank, 2012, “Poverty and Social Impact Analysis Multi-Donor Trust Fund Annual Report, July 2011–June 2012� (Washington, DC: World Bank), 7. 3 Poverty and Social Impact Analysis entry point for the integration of gender into DPOs. The guidance note at hand offers more detailed directions on how to conduct gender-sensitive PSIA, which can then provide an entry point for integrating gender into DPOs. This note provides practical guidance on how to incorporate a gender lens in PSIA. It is not intended as guidance for undertaking separate PSIA focused on gender only. Rather, it is recommended to integrate gender into any PSIA as applicable. The note has three components. First, it provides guiding questions to help assess the need to integrate gender analysis into any specific PSIA. Second, the note uses examples to highlight elements of PSIA into which gender analysis can be integrated, even though many previous PSIA are not gender-informed and the pool from which to draw exam- ples is considerably small. Finally, the note provides some guidance on the methods, data, and tools that can be used to conduct gender-informed PSIA. 1. Is gender important to consider in a PSIA? The gender impact and context of each reform is different and the scope and depth of gender analysis, like any element of PSIA, will vary. Since gender is a cross-cutting social category, it is unlikely that a reform or policy will be entirely gen- der neutral. A gender approach can be absolutely essential and rewarding in some contexts, but more subtle or even impractical in others. Also, depending on the policy or reform itself, the range of issues to be analyzed and the required detail of analysis can vary. The following questions may help task leaders decide whether gender is im- portant to consider in their PSIA. At which level of society are gendered impacts likely to oc- cur? Gendered impacts of policy reform can occur at different levels of society. For example, at the national level, policy reform can affect informal institutions, the pre- vailing norms regarding gender relations and the roles of men and women. In addition, the relative positions of men and women in formal institutions (e.g., markets, govern- ment, legal systems) can both influence and be affected by the reform. At the sector level, economic opportunities, constraints, and incentives for men and women in the “ sector under reform can be affected differently. At the household level, the reform can change the position of men and women within the household, for example, by broad- ening or restricting bargaining power over household deci- It is unlikely that a reform or sions or access to household resources. This, in turn, may “ have implications for individuals because the well-being of policy will be entirely men, women, and other household members who depend on women can be affected. Other effects may also occur at the gender neutral community level. It is often the case that gender impacts take place on multiple levels. Loss of a job and earnings by women, due to the contraction of a sector where women are em- ployed, for example, may well translate into less bargaining power for women in the household, as well as lower investments in the health and education of children. Think- ing through different levels, task teams can brainstorm whether men and women will be affected by the reform differently, and how. 4 Guidance Note Are men and women likely to be affected directly or indirect- ly? Different groups of men and women can be directly affected by a reform, de- pending on how they are engaged with the sector, service, or domain under re- form. Direct gender impacts are relatively easy to identify by disaggregating the im- pacts by gender. For example, in reforms of public sector downsizing and trade liberal- ization, women in the public sector can be affected disproportionately if they consti- tute the majority of the public-sector labor force or other sectors that shrink due to changing import and export patterns. Similarly, energy, water, and natural resource management reforms often have direct welfare consequences on women’s health, time use, and investments in productive skills, compared to men, since women are usually responsible for collecting fuel or fetching water for household use. However, different groups of men and women can also be affected indirectly by the spillover effects or externalities of a reform. Such second- or third-order im- pacts result through channels other than the actual policy lever or action, which makes them somewhat more difficult to identify than the direct effects. The PSIA on trade reforms in Uganda³, for instance, found that increased household incomes due to ex- port growth mainly benefited male members, who recorded higher levels of leisure spending than before. This raised concerns about the potential impacts on the re- sources available for the well-being of other household members, particularly women and children. Men and women can both be targeted users of a particular service that is affected by a reform, but they may face significant differences in accessing it, leading to indirect reform impacts on them. The PSIA on the removal of fuel subsidies in Mozambique� showed that, in terms of direct effects, women were negatively affected as much as men by the rise in transportation cost. However, increased transportation costs also reduced women’s ability to access health facilities, which had longer-term implications on the health care of women and children. Direct and indirect impacts on men and women can also work in tandem. A posi- tive welfare effect can be offset by a negative spillover (and vice versa), or even worse, negative impacts can be exacerbated by the combined effects. A PSIA on mining sector reforms in Romania�, for example, showed that retrenchments affected women dispro- portionately more, but not only because of job losses (direct effect). In addition, wom- en became victims of increased domestic violence and divorce rates (indirect effects). Do men and women face different constraints that prevent them from benefiting from a reform? The welfare impacts of some reforms may seem gender neutral, but social factors can systematically prevent one group from benefiting from the policy change. For example, a subsidy on agricultural inputs to promote the use of modern farming techniques and increase productivity may be ineffective if women, who are usually responsible for production, cannot access it because they cannot travel to buy the in- put. Similarly, urban transport systems designed to transport people to and from em- ployment agencies may respond inadequately to the needs of women, who must com- bine income-generation activities with such tasks as visiting the market and taking children to school. 5 Poverty and Social Impact Analysis Access to decision making about reforms is a particularly important domain where women and men may face significantly different constraints. Poor house- holds are often highly dependent on natural resources for survival and livelihoods, and natural resource collection tasks often rest with girls and women, who bear the opportunity cost of time lost for education and skill development, not to mention poor health and loss in productivity. However, women are often not represented in environmental governance processes; in other words, their concerns as the primary beneficiaries are not included in policy design or implementation. Similarly, women’s voices are often unheard in labor unions, especially in large utility sectors dominated by males. This increases the likelihood of poorly targeted reforms and negative wel- fare impacts, such as higher time or labor costs and gender-differentiated health impacts. 2. Integrating gender analysis into the elements of PSIA Considering men and women as distinct stakeholders Different groups of men and women may be distinct stakeholders in policy re- forms because they have different priorities and are likely to be affected by policies in different ways. Stakeholder analysis is usually conducted as one of the first steps of PSIA. It identifies people, groups, and organizations that are important to take into account when carrying out PSIA. It focuses on who is affected by the policy, as well as who can potentially affect its implementation and results. As benefi- “ ciaries, women and men are often influenced by reforms in different ways. As actors who affect policy implementation (e.g., through civil society In most cases of reform, the organizations, unions, business associations, or industry representatives), men and women represent different view- different interests and points and interests that are not always recognized equally. priorities of groups of men Distinguishing the stakeholders or beneficiaries by gen- “ and women are not equally der often helps reveal different roles of men and wom- en in policy reform. Even when gender-differentiated im- represented pacts are not obvious, considering men and women as sep- arate stakeholders can be useful later in the reform process. The PSIA on the charcoal sector in Tanzania�, for instance, looked at women as distinct stakeholders and identified their interests, as well as potential impacts on them, as the main users of charcoal in households, charcoal sellers, and decision makers for purchasing stoves. In most cases of reform, the different interests and priorities of groups of men and women are not equally represented. A common example is reform in large infrastructure sectors that are dominated by male workers, such as transport and mining. Since women are typically a minority in these labor forces, their specific con- cerns (e.g., safety, health and childcare services at mining sites) are often neglected during consultations with project designers or private companies. On the other hand, depending on the gender balance among the influential actors, gender-informed reforms can encounter opposition and be neglected, reducing the effectiveness of 6 Guidance Note the reform. Thus, it is useful to evaluate how organized men and women are in the sector undergoing reform, identify the differences in the interests of the two groups, and include them in the reform process as separate stakeholders. Identifying transmission channels for gender impacts of poli- cy reforms Transmission channels for policy impacts can be considered from a gender per- spective. After establishing the necessity of considering the gender impacts of the reform, and identifying men and women as separate stakeholders, the most important step in PSIA is delineating the channels by which a particular policy change is expected to impact men and women. Some of the most common transmission channels consid- ered in design, as listed in the PSIA user’s guide�, are employment, prices, access to goods and services, assets, and transfers and taxes. The differentiated impacts on men and women can differ in their timing, by materializing in the short or the long run. They can occur simultaneously or in sequence. Also, the channels at work can be dif- ferent or the same for men and women. In all cases, the ultimate impacts on the two groups can be drastically different. Employment The employment channel can work differently for men and women because they have significantly different positions in the labor market. Women’s time con- straints—due to household, care, and childrearing responsibilities; lower human capital endowments; and additional constraints from markets, cultural norms, and formal or informal institutions—often result in less participation by women in labor markets and lower earnings. In addition, men and women workers tend to cluster in different sec- tors and occupations, a pattern that sometimes overlaps with informal and temporary forms of employment, and pro- or counter-cyclical changes in employment. These differences between men and women in labor force participation, employment, and earnings can both impact the welfare implications of reforms and be affected by the reform. The PSIA for the water sector reform in Malawi� offers an interesting case study. Geared at improving water supply in low-income areas in two cities in Malawi, the PSIA recommended that private kiosks selling water in the cities be regulated. In addition to helping improve access to water by women, the move was seen as provid- ing women with more employment opportunities, since most domestic resellers at private kiosks in the pre-reform scenario were women. The following questions can help evaluate whether employment is a relevant transmis- sion channel through which a policy reform may affect women and men:  Does the reform affect gender earning gaps through employment?  Does the reform affect the employment ratio between men and women?  What are the consequences of occupational segregation for the impacts of the reform?  Are men or women disproportionately employed in the sector of the reform?  Are men or women employed in complementary or substitute sectors of the re- form sector?  Do women and men have the same skills in the reformed sector, especially if they face more competition in the labor market after the reform? 7 Poverty and Social Impact Analysis  Are men or women constrained in any way from taking advantage of the bene- fits of the reform that accrue through employment, such as constraints from household responsibilities, lack of access to resources, limited role in decision making, or others?  Does the reform have an impact on the quality of jobs for men and women?  Does the reform create incentives that deter men and women from taking up paid work in the formal economy?  If retrenchments are part of the reform, how do they affect men and women as employees?  Will changes in income affect investments in the education and health of boys and girls? Prices Changes in the prices of goods or services can affect men and women different- ly, depending on the role of the two groups as producers or consumers in the affected sector or as members of affected households. Different expenditure pat- terns of men and women—resulting from differences in their priorities, needs, and socially constructed responsibilities, for instance—imply that price changes affect the two groups differently. The PSIA undertaken to explore the impact of a proposed water sector privatization in Albania� found that women were willing to pay higher amounts for a regular supply of water and an improved sewerage network than men because the improvements would save them time and allow them to work. In Yemen, the PSIA on the impact of the government’s proposal to eliminate fuel subsidies¹� found that the aggregate effect would be regressive; in other words, energy expendi- tures by the poor could rise by a greater proportion than for the nonpoor. The im- mediate consequence was predicted to be a shift to biomass consumption. To the extent that women are involved in biomass collection and cooking, the PSIA predict- ed that eliminating the subsidy would impose disproportionate health and time costs on women, relative to men. Similarly, changes in the prices of factors of production (wages or inputs) in- duced by a policy reform can have differential impacts on men and women. As discussed before, import competition in female-dominated sectors may lower wom- “ en’s wages, not just for that sector, but also for sectors using the same type of labor skills. If women have less flexibility in changing jobs or sec- tors because they lack the skills or face glass ceilings, then Changes in prices of goods or the effect on wages may be even greater. On the other “ services can affect men and hand, a change in the cost of loans for small and medium enterprises may affect women more adversely than men, women differently since women are overrepresented in the ownership of these enterprises. The following questions can help evaluate whether prices are a relevant transmission channel through which a policy reform may affect women and men differently:  Are men and women likely to respond to a change in the price of a factor of production, a good, or a service differently?  Do men and women have different consumption patterns because of differences in their priorities, needs, and available substitutes? 8 Guidance Note  Does a price change of a good or service have an impact on household members other than the male head (e.g., female members, children, elderly) in terms of time use, household work, and access to health and education, among others? Access to goods and services Men and women have different access to goods and services, and reforms can “ induce behavioral responses that affect their well-being and engagement in eco- nomic activities through this channel. PSIA of infrastructure reforms note the impact of better access to electricity and trans- portation on women’s safety and mobility through street lighting Having access to infrastruc- and transport facilities, respectively. Having access to infrastruc- ture, such as roads, water, and electricity, typically increases ture increases women’s “ women’s participation in agricultural production, in addition to participation in agricultural increasing leisure time or reducing time spent in home produc- tion chores. On the other hand, a reform that does not consider production women’s lack of access to agricultural inputs and financial ser- vices may result in less participation by women in agriculture and lower productivity among female farmers. The following questions can help evaluate whether access to goods and services is a relevant transmission channel through which a policy reform may have different im- pacts on women and men:  Do men or women comprise a major share of the population using the service affected by the reform?  Are vulnerable households, including female-headed households, dependent on access to the goods and services affected by the reform?  If fees are introduced or removed for certain services, are they expected to en- courage or discourage men and women from using these services, and what effect will this have on gender relations?  Will the reform change men’s and women’s access to infrastructure facilities and natural resources and how may these changes affect gender relations?  If industries are closed down, will men and women be affected by the removal of affiliated social services (e.g., child care, health, education)? Assets Men and women usually do not have the same level of control over productive and nonproductive assets. Control of productive assets and reforms affecting it will affect not only productivity but also participation in income-generating activities, espe- cially for women. Lack of regulation that secures women’s ownership of land, for in- stance, may leave women with lower quality land to cultivate (e.g., poorer soil quality, greater distance from markets and water sources), resulting in higher labor input to get the same output, and additional burdens to household chores. Policy changes that impact women’s access to nonproductive assets are likely to have effects on women’s productivity, given that nonproductive assets serve as collateral to access credit and determine the bargaining power within a household. Control over nonproductive re- sources often improves a woman’s fallback position in intra-household bargaining, with implications for decisions on household expenditure and the well-being of wom- 9 Poverty and Social Impact Analysis en and children. The following questions can help evaluate whether assets are a relevant transmission channel through which a policy reform affects women and men differently:  Does the reform facilitate or impede women’s access, ownership, and control over assets?  Does the reform affect women’s ability to accumulate wealth?  Does constrained control over assets prohibit women from benefiting from the reform?  Does the reform affect assets that are disproportionately owned and accessed by women? Transfers and taxes Tax legislation and reforms can have explicit and implicit gender biases. Explicit biases occur when tax deductions or exemptions are openly linked to the gender of the income earner or the primary consumer of a service, which makes them some- what easy to identify. For example, it is common for tax deductions and transfer pay- ments to be allocated to a particular member of the household, usually the male head or the breadwinner. Implicit biases in taxation interact with gender differences in types of employment, consumption behavior, and property rights, leading to dif- ferent impacts. For instance, women predominate in unpaid care work and informal employment, so they may be less affected by direct income taxes than men. Howev- er, women also tend to spend a larger portion of their income on household items, such as food, clothing and health care, which means that an increase in taxes on the- se goods and services has a greater impact on them and their dependents than on men. On the other hand, to the extent that women’s labor supply is more elastic than men’s and the distribution of earnings for women is more concentrated at low levels, taxation of labor income can induce more significant behavior responses by women. Tax and filing systems can create incentives that prevent men and women from tak- ing up paid work in the formal economy, which can reduce the tax base, expand in- formal employment, and produce a variety of impacts on well-being at the house- hold level. The following questions can provide guidance in identifying potential gender im- pacts of tax reforms:  Are women or men overrepresented in certain jobs that are affected by the tax reform?  How are individual work patterns within the household affected by a tax reform?  Are prices of any goods or services purchased by men or women affected?  Do taxes create incentives that deter men or women from taking up paid work in the formal economy?  Do reforms include any tax simplifications limiting deductions and exemptions that affect mostly one gender, such as a limit on childcare deductions? 10 Guidance Note Considering the roles of men and women in institutional analysis Social and market institutions, both formal and informal, can affect the impact of a reform on men and women, given the different roles and constraints the two groups face in these areas. While women’s constraints, compared to men’s, in access- ing the market and other formal institutions, such as the government and legal sys- tems, are well documented, analyzing the role informal institutions and social norms play is also fundamental to understanding how men and women negotiate the reform and gain from it. As mentioned before, women have different networks than men and may not be equally represented in interest groups in the sectors affected by the re- form. Also, in most countries, ministries that concentrate on women’s issues, have less power in the government structure and usually depend on coordination with other ministries to run their programs. In informal governance processes, such as participa- tory budgeting or community consultations, women may not have an equal role or voice. These are factors that affect the ability of men and women to claim benefits of a reform and influence its design. PSIA can include a component on institutional analysis that maps the gender dis- tribution and relations in both formal and informal institutions. This can focus on identifying which formal and informal institutions (including cultural norms) increase or dilute the impact of the reform on men and women, and how. It can also assess what opportunities exist for men and women to play an active role in the institutions through which the reform will be implemented, such as participatory consultations. 3. Methods and tools for gender analysis in PSIA The gender impacts of reforms can be analyzed using quantitative or qualitative approaches, but they are best captured using a combination of the two. While quantitative welfare analysis is generally robust, precise, and representative of the im- pacted populations, the indicators and tools used in welfare analysis usually do not allow gender-disaggregated and intra-household analyses of reform impacts, let alone capture the multidimensionality of gender or the cultural specificities of gender rela- tions. Qualitative approaches are typically better equipped to analyze social relations, but are not as representative and robust as quantitative ones. Combining qualitative and quantitative information and techniques can go a long way in making up for the individual shortcomings of each method and its data types, in regards to gender analy- sis. Box 3 gives an overview of some resources that provide guidance on the analysis of quantitative and qualitative information on gender impacts of policy reforms. Quantitative approaches are widely used in PSIA to assess the welfare impacts of reforms on the affected populations, including men and women. A common way of integrating gender into PSIA is to disaggregate survey data by the gender of the household head to show differences, if any, between female- and male-headed house- holds. The PSIA on fuel tax reform in Mozambique¹¹ included simulations of changes in fuel prices by the gender of the household head. The analysis revealed differences in the impact of increasing the fuel tax and female-headed households emerged as par- ticularly vulnerable. In other cases, PSIA uses individual-level data disaggregated by gender. In Uruguay, the PSIA on tax reform¹² analyzed the impact of eliminating the 11 Poverty and Social Impact Analysis wage tax and introducing a dual personal income tax on the labor income of both men and women. Another PSIA looking at the impact of a rise in school fees in Mozambique¹³ gathered separate data on women, their school enrollment, and edu- cation level completed, and how a fee increase would affect these elements. Qualitative approaches can play a crucial role to capture certain aspects of gen- der relations. They are better suited for developing an understanding of processes and perceptions of gender and gender relations, which is often essential to under- standing different reform impacts on women and men. If a purely quantitative ap- proach is constrained by the unavailability of sufficiently disaggregated data (particularly, intra-household level or individual data), qualitative approaches provide an alternative solution. Focus group discussions, for instance, are useful for capturing the voices and priorities of men and women. The PSIA on retrenchments in the min- ing sector in the Democratic Republic of Congo¹� conducted focus group discussions with wives of workers participating in the voluntary departure program. In addition, interviews with key informants, including women’s organizations, on women’s con- cerns in the reform sector, can also provide important information. Box 3. Resources on the analysis of gender across sectors These publications include a range of tools for economic, social, and political analysis of reforms, some of which can integrate a gender lens:  World Bank, 2003, “A User’s Guide to Poverty and Social Impact Analysis� (Washington, DC: World Bank).  World Bank, 2007, “Tools for Institutional, Political, and Social Analysis of Policy Reform – A Sourcebook for Development Practitioners� (Washington, DC: World Bank). These are tools, sourcebooks, and other resources for gender analysis that can be integrated into PSIA:  ADePT Gender. This tool produces tables and graphs using household surveys to help diagnose and analyze gender inequalities. It is available online under www.worldbank.org under Research – Analytical tools.  World Bank, 2005, “Gender Responsive Social Analysis: A Guidance Note. Incorporating Social Dimensions into Bank-Supported Projects� (Washington, DC: World Bank).  World Bank, 2005, “Gender Issues in Monitoring and Evaluation in Rural Development: A Tool Kit� (Washington, DC: World Bank).  M. Bamberger, M. Blackden, L. Fort, and V. Manoukian, 2002, “Gender: A Sourcebook for Poverty Reduction Strategies�, ed. Jeni Klugman, chapter 10 pp. 333-74 (Washington, DC: World Bank).  World Bank, FAO, and IFAD, 2000, “Gender in Agriculture Sourcebook� (Washington, DC: World Bank).  World Bank, 2012, “Good Practice Note: Integrating Gender into Development Policy Loans � (Washington, DC: World Bank). Currently only available to World Bank staff.  World Bank, n.d., “Gender Informed Monitoring and Evaluation in Disaster Risk Management. East Asia and Pacific Region Sustainable Development Gender and Disaster Risk Management Guidance Note 3� (Washington, DC: World Bank).  World Bank, 2009, “Improving Gender Targeting of Public Expenditures – A Guidance Note on How to Ad- dress Gender Considerations in Public Expenditure Reviews� (Washington, DC: World Bank). 12 Guidance Note 4. Data for gender analysis in PSIA Different data sources, including administrative data, house- Box 4. Gender equality data and statis- hold surveys, and regional and national aggregates, can be tics combined when analyzing the impact of a reform. While admin- istrative data and national aggregates can provide information of This portal is a one-stop shop for gender the changes introduced by the reform, microdata (e.g., household information, catering to a wide range of surveys, qualitative data, etc.) are best suited to measure the im- users: pacts by gender. For tax reform, for example, national and regional  The eAtlas of Gender allows users to aggregates, such as prices, provide information on the effect of the map and graph dozens of gender reform. In a reform affecting the provision of a service, administra- indicators over time and across tive data can be used to measure changes in the use of the service, countries. even disaggregating by large groups. Household data can also be  The Enterprise Surveys database pro- used to estimate the impact of a reform. For example, the PSIA vides firm-level data on women’s conducted in Mozambique¹� to analyze the impact of higher entrepreneurship and participation school fees can combine administrative data coming from schools, in the labor force. as well as household data, to understand more deeply the impacts on the affected children, other family members, and other house-  A comprehensive list of 52 core indi- holds. It is essential to determine early on in the process whether cators agreed upon by the Inter- the available data, its frequency, and other characteristics are suffi- Agency Expert Group on Gender cient to analyze the policy change from a gender perspective or Statistics can be used to cover sever- whether new data needs to be collected. In some cases, for exam- al dimensions of impact. ple, the years for which data was collected before and after the  The Key Gender-Relevant Employ- policy reform are too distant in time and new data must be gath- ment Indicators provide data on la- ered. bor markets  Findex provides data on access to A range of indicators capturing income and nonincome dimen- finance. sions can provide insights into the gender impacts of policy reforms. Key sources for gender-informed data are listed in box 4.  Regional data provides access to The data can include standard measures of socioeconomic well- major sources of global, regional, being and demographic characteristics, as well as nonstandard and national statistics across UN and measures, such as time use, safety, aspirations, intra-household other agencies and regional com- relations, and perceptions of well-being. Reforms affecting the missions and World Bank regions. provision or cost of a service can also include indicators, such as  Women, Business, and the Law pre- sents indicators based on laws and intra-household differences in consumption of the service under regulations affecting women’s pro- reform, women’s perception on the quality of service, costs in- spects as entrepreneurs and em- curred because of quality deficiencies, time spent in accessing the ployees in 128 economies across 6 service, price of service and men and women’s willingness to pay, topics. and alternative services used by women and their prices, and how these elements impact on health, education, safety, and intra- The portal can be accessed online at marital relations, among others. While specific indicators may de- http://datatopics.worldbank.org/gender pend on the sector under consideration and the sociocultural con- text of the country or region in which the reform takes place, it is also advisable to go beyond gender gaps on monetary and non- monetary measures and consider indicators that reflect women’s lack of access to resources, unequal legislation, and other custom- ary laws and gender-specific constraints. For land policy reforms, for instance, it may be useful to include a gender differentiated indicator on access to agricultural inputs such as credit. 13 Poverty and Social Impact Analysis 5. Conclusion This guide shows how a gender lens can be integrated into PSIA. After assessing the need to integrate gender analysis into any specific PSIA, task teams have a varie- ty of options for conducting this type of analysis. Existing data sources capture a range of helpful gender indicators and further data can be collected for qualitative and quantitative analysis. By informing DPOs and policy reforms in general, gender- informed PSIA can help improve both reform outcomes and gender equality. Integrating gender into PSIA that is being conducted in data-scarce situations can be particularly challenging. Gender issues are often the first to be dropped if data availability is limited, especially if teams are also short on time and resources. This, however, can mean that gender impacts of policy reforms are neglected in situ- ations where analysis may be particularly needed. Under such circumstances, PSIA task teams may find it helpful to draw on existing evidence and intense consultations with key stakeholders to compensate for a lack of original and full-fledged data col- lection and analysis. For more information, visit the Poverty and Social Impact Analysis website: http:.//www.worldbank.org/psia 14 Guidance Note Notes 1. World Development Report 2012 argues that outcomes for women can improve through accumulation of endowments (education, health, and physical assets), the use of such endowments to take advantage of economic opportunities to generate income, and the application of endowments to take action or improve agency. See World Bank, 2011, “World Development Report 2012: Gender Equality and Development� (Washington, DC: World Bank). 2. The latest DPO retrospective (covering FY 09–12) indicates an upward trend in the number of PSIA completed to inform DPOs. It showed that, of the 221 DPOs included in the retrospective, about half (46 percent) contain one or more prior actions with poten- tially negative short-term distributional impacts. Prior actions with potentially negative short-term distributional impacts require PSIA, as stipulated in OP 8.60. Of a total of 2,186 prior actions, about 8 percent had potentially negative distributional impacts. Of the 8 percent of prior actions which needed PSIA, it was conducted in 71 percent of the cases. 3. D. Booth, D. Kasente, G. Mavrotas, G. Mugambe, and A. Muwonge, 2003, “The Strategic Exports Initiative in Uganda� (Washington, DC: World Bank). 4. K. Nicholson, B. O’Laughlin, A. Francisco, and V. Nhate, 2003, “Fuel Tax in Mozambique� (Washington, DC: World Bank). Currently only available to World Bank staff. 5. World Bank, 2005, “Poverty and Social Impact Analysis of Mining Sector Reform in Romania: A Policy Note� (Washington, DC: World Bank). 6. World Bank, 2010, “Enabling Reforms: A Stakeholder-Based Analysis of the Political Economy of Tanzania’s Charcoal Sector and the Poverty and Social Impacts of Proposed Reforms� (Washington, DC: World Bank). 7. World Bank, 2003, “A User’s Guide to Poverty and Social Impact Analysis� (Washington, DC: World Bank). 8. E. Chirwa and N. Junge, 2007, “Private Sector Participation in the Distribution and Management of Water Services: A Study of Low- Income Areas in Blantyre and Lilongwe� (Lilongwe, Republic of Malawi: Ministry of Economic Planning and Development, Ministry of Irrigation and Water Development). 9. S. Beddies, H. De Soto, A. Bakllamja, X. Chauvot de Beauchene, 2004, “Decentralization and Water Sector Privatization in Alba- nia� (Tirana: World Bank and United Nations Development Program). Currently only available to World Bank staff. 10. World Bank and United Nations Development Programme, 2005, “Household Energy Household Energy Supply and Use in Yem- en� (Washington, DC: World Bank, Energy Sector Management Assistance Program). 11. K. Nicholson, B. O’Laughlin, A. Francisco, and V. Nhate, 2003, “Fuel Tax in Mozambique� (Washington, DC: World Bank). Currently only available to World Bank staff. 12. World Bank, 2005, “Uruguay Poverty and Social Impact Assessment of the Tax Reform� (Washington, DC: World Bank). 13. World Bank, 2005, “Mozambique Poverty and Social Impact Analysis: Primary School Enrollment and Retention—The Impact of School Fees� (Washington, DC: World Bank). 14. World Bank, 2007, “Democratic Republic of the Congo Poverty and Social Impact Analysis: Mine Sector Reform� (Washington, DC: World Bank). 15. World Bank, 2005, “Mozambique Poverty and Social Impact Analysis: Primary School Enrollment and Retention—The Impact of School Fees� (Washington, DC: World Bank). Acknowledgments This guidance note has been prepared by Lisa Schmidt, Emcet Tas, Josefina Posadas, Gisela Garcia, Maria Beatriz Orlando and Soumya Kapoor under the guidance of Maitreyi Bordia Das. Comments from Jeni Klugman, Vanessa Lopes Janik, Elizaveta Perova, Jonna Lundvall and Miriam Muller are gratefully acknowledged. 15 The World Bank Social Development Department 1818 H Street NW Gender and Development Group Washington DC www.worldbank.org