Impact of social protection on gender equality in low‐ and middle‐income countries: A systematic review of reviews

BACKGROUND: More than half of the global population is not effectively covered by any type of social protection benefit and women's coverage lags behind. Most girls and boys living in low‐resource settings have no effective social protection coverage. Interest in these essential programmes in low and middle‐income settings is rising and in the context of the COVID‐19 pandemic the value of social protection for all has been undoubtedly confirmed. However, evidence on whether the impact of different social protection programmes (social assistance, social insurance and social care services and labour market programmes) differs by gender has not been consistently analysed. Evidence is needed on the structural and contextual factors that determine differential impacts. Questions remain as to whether programme outcomes vary according to intervention implementation and design. OBJECTIVES: This systematic review aims to collect, appraise, and synthesise the evidence from available systematic reviews on the differential gender impacts of social protection programmes in low and middle‐income countries. It answers the following questions: 1. What is known from systematic reviews on the gender‐differentiated impacts of social protection programmes in low and middle‐income countries? 2. What is known from systematic reviews about the factors that determine these gender‐differentiated impacts? 3. What is known from existing systematic reviews about design and implementation features of social protection programmes and their association with gender outcomes? SEARCH METHODS: We searched for published and grey literature from 19 bibliographic databases and libraries. The search techniques used were subject searching, reference list checking, citation searching and expert consultations. All searches were conducted between 10 February and 1 March 2021 to retrieve systematic reviews published within the last 10 years with no language restrictions. SELECTION CRITERIA: We included systematic reviews that synthesised evidence from qualitative, quantitative or mixed‐methods studies and analysed the outcomes of social protection programmes on women, men, girls, and boys with no age restrictions. The reviews included investigated one or more types of social protection programmes in low and middle‐income countries. We included systematic reviews that investigated the effects of social protection interventions on any outcomes within any of the following six core outcome areas of gender equality: economic security and empowerment, health, education, mental health and psychosocial wellbeing, safety and protection and voice and agency. DATA COLLECTION AND ANALYSIS: A total of 6265 records were identified. After removing duplicates, 5250 records were screened independently and simultaneously by two reviewers based on title and abstract and 298 full texts were assessed for eligibility. Another 48 records, identified through the initial scoping exercise, consultations with experts and citation searching, were also screened. The review includes 70 high to moderate quality systematic reviews, representing a total of 3289 studies from 121 countries. We extracted data on the following areas of interest: population, intervention, methodology, quality appraisal, and findings for each research question. We also extracted the pooled effect sizes of gender equality outcomes of meta‐analyses. The methodological quality of the included systematic reviews was assessed, and framework synthesis was used as the synthesis method. To estimate the degree of overlap, we created citation matrices and calculated the corrected covered area. MAIN RESULTS: Most reviews examined more than one type of social protection programme. The majority investigated social assistance programmes (77%, N = 54), 40% (N = 28) examined labour market programmes, 11% (N = 8) focused on social insurance interventions and 9% (N = 6) analysed social care interventions. Health was the most researched (e.g., maternal health; 70%, N = 49) outcome area, followed by economic security and empowerment (e.g., savings; 39%, N = 27) and education (e.g., school enrolment and attendance; 24%, N = 17). Five key findings were consistent across intervention and outcomes areas: (1) Although pre‐existing gender differences should be considered, social protection programmes tend to report higher impacts on women and girls in comparison to men and boys; (2) Women are more likely to save, invest and share the benefits of social protection but lack of family support is a key barrier to their participation and retention in programmes; (3) Social protection programmes with explicit objectives tend to demonstrate higher effects in comparison to social protection programmes without broad objectives; (4) While no reviews point to negative impacts of social protection programmes on women or men, adverse and unintended outcomes have been attributed to design and implementation features. However, there are no one‐size‐fits‐all approaches to design and implementation of social protection programmes and these features need to be gender‐responsive and adapted; and (5) Direct investment in individuals and families' needs to be accompanied by efforts to strengthen health, education, and child protection systems. Social assistance programmes may increase labour participation, savings, investments, the utilisation of health care services and contraception use among women, school enrolment among boys and girls and school attendance among girls. They reduce unintended pregnancies among young women, risky sexual behaviour, and symptoms of sexually transmitted infections among women. Social insurance programmes increase the utilisation of sexual, reproductive, and maternal health services, and knowledge of reproductive health; improve changes in attitudes towards family planning; increase rates of inclusive and early initiation of breastfeeding and decrease poor physical wellbeing among mothers. Labour market programmes increase labour participation among women receiving benefits, savings, ownership of assets, and earning capacity among young women. They improve knowledge and attitudes towards sexually transmitted infections, increase self‐reported condom use among boys and girls, increase child nutrition and overall household dietary intake, improve subjective wellbeing among women. Evidence on the impact of social care programmes on gender equality outcomes is needed. AUTHORS' CONCLUSIONS: Although effectiveness gaps remain, current programmatic interests are not matched by a rigorous evidence base demonstrating how to appropriately design and implement social protection interventions. Advancing current knowledge of gender‐responsive social protection entails moving beyond effectiveness studies to test packages or combinations of design and implementation features that determine the impact of these interventions on gender equality. Systematic reviews investigating the impact of social care programmes, old age pensions and parental leave on gender equality outcomes in low and middle‐income settings are needed. Voice and agency and mental health and psychosocial wellbeing remain under‐researched gender equality outcome areas.