ENERGY VULNERABILITY IN FEMALE-HEADED HOUSEHOLDS Findings from the Listening to Citizens of Uzbekistan Survey DECEMBER 2019 © 2019 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of the World Bank, with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because the World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202- 522-2625; e-mail: pubrights@worldbank.org. Cover photo: © World Bank. Further permission required for reuse. ENERGY VULNERABILITY IN FEMALE-HEADED HOUSEHOLDS Findings from the Listening to Citizens of Uzbekistan Survey CONTENTS ACKNOWLEDGMENTS .....................................................................................................................iv EXECUTIVE SUMMARY ......................................................................................................................v ABBREVIATIONS ................................................................................................................................vi 1. INTRODUCTION ........................................................................................................1 Objectives of the Study...............................................................................................2 Findings from Previous Studies on Gender and Energy Access .........................3 2. HOUSEHOLD SURVEY PROFILE ............................................................................ 6 3. FINDINGS OF THE L2CU SURVEY.......................................................................... 9 Access to (and Quality of) Energy Services ............................................................9 Affordability and Ability to Pay for Utilities ......................................................... 13 Coping Mechanisms ...................................................................................................22 Priorities for Public Services and Infrastructure................................................ 28 4. CONCLUSION.......................................................................................................... 30 APPENDIX A........................................................................................................................................32 REFERENCES .................................................................................................................................... 37 ACKNOWLEDGMENTS A team from the World Bank Group’s Social Development and Energy and Extractives Global Practices (GPs) prepared this report. Ursula Casabonne (Senior Gender Consultant, Social Development GP, Europe and Central Asia (ECA) Region), Maria Manuela Faria (Social Development Specialist, Social Development GP, ECA Region), Ricardo Graiff Garcia (Consultant, Department of Political Science, Ohio State University), and Audrey Sacks (Senior Social Development Specialist, Social Development GP, ECA Region) wrote this report. The report benefitted from input from Mitsunori Motohashi (Senior Energy Specialist, Energy and Extractives GP, East Asia and Pacific Region), and capacity to achieve environmentally sustainable energy Hiwote Tadesse (Operations Officer, Energy and Extractives solutions for poverty reduction and economic growth. ESMAP GP, ECA Region). The team is grateful to Vinayakraj is funded by Australia, Austria, Canada, ClimateWorks Nagaraj (Senior Economist, Macroeconomics, Trade, and Foundation, Denmark, the European Commission, Investment GP, ECA Region), Inka Schomer (Operations Finland, France, Germany, Iceland, Italy, Japan, Lithuania, Officer, Energy and Extractives GP, Global Programs Unit), Luxembourg, the Netherlands, Norway, the Rockefeller William Seitz (Economist, Poverty and Equity GP, ECA Foundation, Sweden, Switzerland, and the United Kingdom, Region), Gulrano Tlepova (Social Development Specialist, as well as by the World Bank. Social Development GP, ECA Region), and Robert Wrobel (Senior Social Development Specialist, Social Development The team is also grateful for the support received from the GP, ECA Region) for their constructive feedback. The report Support Program for Socially, Environmentally, and Financially was completed under the guidance of Kevin Tomlinson Sustainable Production of Cotton in the Republic of Uzbekistan (Practice Manager, Social Development GP, ECA Region). Multi-Donor Trust Fund, which the World Bank established with funding from the European Union, the United States, The team gratefully acknowledges the support from the Switzerland and Germany. The Multi-Donor Trust Fund Energy Sector Management Assistance Program (ESMAP). finances activities to eliminate child labor and forced labor ESMAP is a global knowledge and technical assistance in the cotton sector in Uzbekistan and strengthen the program administered by the World Bank. It provides capacity of state and nonstate institutions to ensure the analytical and advisory services to low- and middle-income sustainability of cotton production and agricultural reform countries to increase their know-how and institutional in general. iv Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey EXECUTIVE SUMMARY Uzbekistan has the second-most-energy-intensive The analysis suggests that FHHs have similar access to economy in Europe and Central Asia as measured according energy as MHHs but find it harder to pay for utilities and to energy intensity per unit of gross domestic product. This basic needs. FHHs are more likely than MHHs to reduce study assesses the sex-disaggregated effect of access their food consumption and borrow money to pay for basic to energy across the country using data from the 2018 needs. Poor FHHs with employed household members are and 2019 waves of the Listening to Citizens of Uzbekistan more likely to reduce their food consumption than MHHs household survey, which is representative at the national, with no employed household members. regional, urban, peri-urban, and rural levels. The analysis focuses on three main factors: the challenges that female-headed households (FHHs)—which constitute 20 percent of the sample—face in accessing reliable electricity and heating services; whether FHHs struggle more than male-headed households (MHHs) to pay for utilities; and how often FHHs use coping mechanisms such as reducing food, health care, and other purchases to meet basic needs. Executive Summary v ABBREVIATIONS ECA Europe and Central Asia ESMAP Energy Sector Management Assistance Program FHH Female-headed household GDP Gross domestic product GP Global practice L2CU Listening to Citizens of Uzbekistan MHH Male-headed household UZS Uzbekistan som vi Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey 1. INTRODUCTION Uzbekistan has the second-most-energy-intensive economy in Europe and Central Asia as measured according to energy intensity per unit of gross domestic product (GDP). Although its energy intensity declined by approximately 68 percent from 1998 to 2013, its energy “... households headed by use per unit of GDP is 2.1 times as high as the regional average (World Bank 2013). The residential sector accounts women and men may differ in for 40 percent of total energy consumption (UNDP 2016). their level of access, ability Three types of centralized energy services are available to households, small enterprises, and public buildings in to pay for energy services, Uzbekistan: central heating systems, including domestic energy needs and uses, and hot water; electricity; and gas. Access to these services is not universal. Of the households surveyed for the Listening understanding of energy to Citizens of Uzbekistan (L2CU) survey, 97 percent have access to electricity, 47 percent have access to centralized efficiency improvements.” gas (Although only 22 percent of the households surveyed are located in urban areas, 40 percent of the surveyed households that have access to centralized gas are urban.), and 6 percent have access to central heating (nearly all urban) (figure 1). Central heating systems installed between the 1950s and Figure 1.  Access to Centralized Energy Services the 1970s were designed as open systems for hot water supply, in which water from the central heating network was transferred directly to the domestic hot water system and radiators in public buildings. This design accelerated deterioration of heat transportation and distribution networks and led to inefficient energy use. Underinvestment in maintenance, rehabilitation, and modernization of the district’s heating systems over the past two decades, which has resulted in further deterioration of its heating service, has compounded these deficiencies. Central heating services have degraded in all cities of Uzbekistan; in several cities, the services have stopped working for parts or all of the city. Where the system has failed, electricity and gas have been used extensively to provide heat and hot water in multiapartment buildings and public buildings, causing several problems, including inefficient use of electricity and natural gas; overloaded power sector infrastructure, which accelerates deterioration and thus leads to frequent power outages, especially in winter; and safety and health risks caused by the use of poor-quality or polluting heaters and stoves (World Bank 2018). Source: L2CU Household Survey (June/July 2018). Uzbekistan’s President, Shavkat Mirziyoyev, elected in Note: Figure 1 shows the percentage of households with access to centralized energy services. Access to electricity is nearly universal 2016, launched a large-scale reform program to increase for Listening to Citizens of Uzbekistan (L2CU) households. Access access to good-quality public services in dialogue with to central heating is rare; nearly every household with access to citizens to ensure greater accountability and transparency central heating is urban. Access to centralized gas is more common, particularly for urban households. and to improve the population’s well-being. A major Introduction 1 element of this program involves increasing access to centralized energy services with alternative nonnetworked energy resources and increasing the efficiency and financial sources. Respondents expressed a strong desire to improve sustainability of the energy sector. Related policy measures their situation. Some communities are paying for the costs include modernization, upgrades, and renovation of the of maintenance and repairs themselves and are willing to pay energy services infrastructure, especially in rural areas; more for higher-quality services if reforms are implemented introduction of energy-efficient technologies; and increases (Hiller et al. 2016). in energy tariffs to improve the financial returns of energy services. The conventional approach to energy policy and planning is based on the assumption that a good energy policy, BOX 1. Listening to Citizens of Uzbekistan program, or project will meet the practical needs of men (L2CU) Survey and women equally, but households headed by women and men may differ in their level of access, ability to pay for energy services, energy needs and uses, and understanding The L2CU survey, a collaborative effort of the World of energy efficiency improvements. An approach to energy Bank and the Development Strategy Center in policy and planning that overlooks the differences between Tashkent, was created to comprehensively monitor male-headed households (MHHs) and female-headed the views and well-being of a representative group households (FHHs) could have unintended differential effects of people during introduction of policy reforms. on and benefits for men and women. The L2CU comprises a 4,017-household nationally representative baseline survey, a monthly panel survey of a subset of 1,500 households from the Objectives of the Study baseline survey, and qualitative data collection from specialists and focus groups. The baseline survey was Reliable data and evidence on access to (and quality of) conducted in June and July 2018, and the monthly energy services and the differential effects of energy panel surveys were initiated in September 2018 and policies on households’ welfare has only recently become are expected to continue for the foreseeable future. available. The L2CU survey gathers monthly data on access to public services (including energy) and other topics such as The surveys address topics related to poverty, employment and citizens’ views on the country’s economic housing, migration, and employment. The questions outlook (box 1). cover income and income sources; shocks and coping strategies; employment; migration; access Using the L2CU baseline and monthly data, this study to public services (including energy); interactions assessed the gender-disaggregated effect of energy with service providers, government representatives, access in Uzbekistan. It focused on three main questions: and Mahalla Citizens Assembly (i.e., community self- governing bodies) representatives; and views on the • Do FHHs face challenges in accessing reliable electricity country’s governance. The qualitative component, and heat? which was launched in November 2018, uses a • Do FHHs struggle more than MHHs to pay for utilities? combination of focus group discussions and key • How often do FHHs use coping measures such as informant interviews to investigate in greater detail reducing food, health care, and other purchases to meet matters related to governance, accountability, basic energy needs? service delivery, and livelihoods. The study complements an earlier qualitative study of By tracking people’s experiences over the course Uzbek households’ experiences with and perceptions of of a year, the study demonstrates how policies poor-quality energy services. Access to centralized energy affect people’s daily lives. An advisory council services and quality is generally poor to moderate in many of representatives of the State Statistical areas outside Tashkent, which causes many households to Committee, government ministries, and national use inefficient—and sometimes unsafe—coping strategies think tanks advised on the study’s design and (Hiller et al. 2016). The monetary and nonmonetary costs provides continuous assistance in the review and of those coping strategies can be high, particularly for rural interpretation of the results. and low-income households, which often accumulate debt to pay for centralized energy services. Furthermore, low- Source: World Bank ND. income households may be unable to supplement poor 2 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Findings from Previous Studies on Gender and in income-earning activities. An unreliable energy supply Energy Access affects women employed in the formal sector in two ways: they must manage their job responsibilities while power is Previous studies have found strong linkages between available, and they must meet household and family needs gender and energy access in six critical areas: access to when they come home from work, in what has been referred (and quality of) energy services; energy affordability, tariffs, to as the “second shift.” and pricing; environment and health and safety (exposure to energy hazards); energy conservation decisions; coping Energy Affordability, Tariffs, and Pricing strategies for poor energy access, affordability, and quality; and user knowledge, access to information, and interaction Low-income and poor households often have a higher with energy providers (Hiller et al. 2016; World Bank 2015a, energy burden than higher-income households because 2016; Ajwad et al. 2014; Canpolat and Georgieva 2019) they spend a greater percentage of their income on energy. (figure 2). They also often live in less-energy-efficient housing and pay more per square foot for energy because they cannot afford Access to (and Quality of) Energy Services improvements that would make their homes more energy efficient; they may also lack information about such options. Access to affordable energy is essential for fulfilling basic Energy constitutes a significant share (10–20 percent) of needs such as cooking, heat, and mobility. Energy scarcity annual household expenditures in Uzbekistan: approximately has a disproportionate effect on women and girls, who rely USD 264 (UZS 750,000) for a family of five (USD 53 (UZS on labor-saving appliances (e.g., washing machines, electric 150,000) per capita) (Hiller et al. 2016). Some low-income stoves, vacuum cleaners) to perform traditional household households reported that their energy expenditures rose chores. Without reliable energy sources, women must to 50 percent of their income during winter (Hiller et al. perform these jobs by hand, which leaves less time to engage 2016). The L2CU findings show annual household spending Figure 2: Critical Gender Equality Aspects of Demand-Side Energy Access Access to (and quality of) energy services User knowledge, Energy affordability, access to information, tariffs, and pricing interaction with energy providers Coping strategies for Environment, health poor energy access and safety (exposure to and quality energy hazards) Energy conservation decisions Introduction 3 of approximately USD 68.95 (UZS 583,500) for a family Energy Conservation Decisions of five (USD 13.79 (UZS 116,700) per capita), although the average L2CU household reports spending only 3 percent to Energy conservation measures and interventions should 4 percent of its annual budget on energy (World Bank ND). focus on women, because they are the predominant users and managers of energy in the household. In Uzbekistan, FHHs could be particularly disadvantaged because they women take more responsibility than men for household tend to be poor households. Women headed almost 20 energy conservation and in educating other members, percent of households in the baseline L2CU survey. This is particularly children, to conserve energy (Hiller et al. 2016). consistent with 2002 Demographic and Health Survey data In Ferghana oblast center, women were found to be more (Republic of Uzbekistan 2004), which report that women familiar with how much energy the household consumes; headed 18 percent of the country’s households. Large-scale where the bill is perceived to be higher than the household’s male outmigration in recent decades (World Bank 2016; consumption, they are not afraid to argue with inspectors. Ajwad et al. 2014) has contributed to the increase in the number of FHHs in Uzbekistan. FHHs are more likely to have lower incomes than MHHs. Among households in the monthly Coping Strategies for Limited Energy Access, L2CU telephone survey, FHHs are generally 30 percent to 50 Affordability, and Quality percent less likely than MHHs to be able to afford utilities or food and 30 percent to 40 percent more likely to report Rising energy expenditures and unreliable energy supply borrowing money to pay for basic household needs. Energy disproportionately affect women and FHHs across affordability for elderly women who live alone on a fixed low socioeconomic groups (Canpolat and Georgieva 2019). They income deserves special consideration in Europe and Central cope by: Asia (World Bank 2015a). Such households may require more social assistance or additional support to ensure that they • Reducing spending on food. Women are more likely to can fulfill basic needs, including heating, utilities, and food. sacrifice their own food consumption first (World Bank Where male migration is high and there are few opportunities 2015a). to earn a cash income locally, such as in Uzbekistan, FHHs • Reducing healthcare expenses by avoiding doctor visits or may also need additional support to pay for energy (World self-treating. Households might stop going to hospitals Bank 2015a). or health clinics, reduce or skip medications, or use traditional treatments. Elderly individuals, who live on smaller, fixed incomes and have higher, more regular Environment and Health and Safety (Exposure to medical costs, most often mention resorting to this Energy Hazards) strategy. • Reducing expenses on education, clothing, social Women are directly exposed and vulnerable to health and gatherings, and traditional celebrations. Households environmental energy-related hazards. According to Global might not be able to afford uniforms or tuition fees Health Observatory data (WHO ND), in 2016, household for better-quality schools. They may stop attending air pollution caused 3.8 million deaths (mainly women and social gatherings and traditional celebrations such as children) from fumes emitted by biomass-based fuels, weddings because of transportation and gift expenses, accounting for 7.7 percent of global mortality. In many which can result in social isolation. regions, such as South Asia and parts of sub-Saharan Africa, • Borrowing money from relatives, friends, and financial it has become the most important risk factor for ill health— institutions. The poorest households may be more likely higher even than unsafe water and sanitation (Dutta et to borrow from relatives or friends. In certain cases, al. 2017). Uzbek focus groups reported several effects of they might borrow money from financial institutions, limited access to energy in Uzbekistan, including greater which can cause stress and impose the additional cost incidence of illness (colds and influenza) in winter and the of interest. inconvenience of the whole family living in the only room that • Reducing household energy use as much as possible. has heating; safety concerns associated with unpredictable To reduce bills, households may reduce their energy centralized gas outages that could cause gas leakages and use by heating only one room in the house, changing explosions; safety concerns for women, children, and elderly their bathing habits, avoiding spending time at home adults due to a lack of street lighting after dark; and greater to save on heating costs, using electrical appliances time burden and costs of using solid fuels (firewood, cotton less, and performing certain tasks such as washing stalks, animal manure) due to the amount of time required clothes manually. Extended families in cold-climate to collect these materials and the high cost of transporting countries may live together in the winter months, often them (Hiller et al. 2016). in crowded conditions, to save on heating expenses. 4 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey • Delaying payments and not paying bills. Households may Qualitative research in Europe and Central Asia suggests be forced to delay bill payments, which often incurs that cultural norms affect the way men and women relate penalties. If they fail to pay their energy bills, they will to energy providers and social assistance offices (World be disconnected. Bank 2015a). A Europe and Central Asia study on the gender- disaggregated effect of energy subsidy reforms found that In Uzbekistan, families cope with low temperatures at home interacting with energy providers (seeking information, in winter by residing in one room; wearing heavy clothing reporting service problems, inquiring about inconsistent bills) inside (including hats and scarves); covering windows, doors, is more often a male responsibility because men are thought and floors with film, cloth, or blankets; using electric ovens to be more authoritative and assertive and better informed as space heaters; moving to temporary living quarters; about the technical aspects of such inquiries (World Bank avoiding using electric devices; and shifting from centralized 2015a). energy services to solid fuels (Hiller et al. 2016). Although some low-income households go into debt to pay for Several women in urban areas reported that they are centralized energy services, others avoid doing so out of fear responsible for routine interactions with providers, such of being disconnected and having to pay a reconnection fee. as paying bills, but that they would send their husbands to resolve a grievance because this is too stressful and time consuming and the results are uncertain (World Bank User Knowledge, Access to Information, and 2015a). The Europe and Central Asia study reveals that Interaction with Energy Providers women who are reluctant to interact with energy providers are also more passive about understanding the specifics Women may lack information about energy efficiency of tariff reforms (World Bank 2015a). In rural areas, where and affordable options. A World Bank (2015a) qualitative traditional gender roles seem stronger, women are often study on energy subsidy reform in eight countries in Europe burdened with collecting solid fuels while men deal with and Central Asia found that men are better informed about energy bill payments, particularly if the inspectors are male appliances’ technical characteristics, energy efficiency, and (Hiller et al. 2016). reliability, so households are likely to defer to men’s decisions. “Female-headed households are generally 30 percent to 50 percent less likely than male-headed households to be able to afford utilities or food and 30 percent to 40 percent more likely to report borrowing money to pay for basic household needs.” Introduction 5 2. HOUSEHOLD SURVEY PROFILE The L2CU representative baseline survey was administered in a FHH; 84.2 percent of MHHs and 76.8 percent of to 4,017 households, of which 20 percent (801 households) FHHs had at least one member younger than 18, and were FHHs and 80 percent (3,216 households) were MHHs. approximately 37 percent of households had at least The surveyed households had the following demographic one grandchild of the head of household living in the characteristics. household (figure 5). • Age of head of household: The mean age of a head of • Sex of survey respondents: 52.8 percent of respondents household was 49.7 for men and 57.0 for women (figure in MHHs that completed the survey were women, and 6). Nearly one-third of female heads of household were 7.9 percent of respondents in FHHs were men (figure 3). 70 or older. • Marital status: 94 percent of male heads of household • Household consumption quintiles: The proportion were married; 75 percent of female heads of household of households in different consumption quintiles was were widowed, and 10 percent were divorced (figure 4). similar for MHHs and FHHs, although FHHs were more • Age of household members: The mean age of a likely to be in the lowest quintile (figure 7). household member was 28.3 years in a MHH and 28.8 Figure 3.  Sex of Survey Respondents, According to Figure 4.  Marital Status of Head of Household Sex of Head of Household Source: L2CU household survey (June/July 2018). Source: L2CU household survey (June/July 2018). Note: Figure 3 shows the percentage of male and female respondents in Note: Figure 4 shows the marital status of male and female heads of male- and female-headed households. The majority of the respondents household. The vast majority of female heads of household are widows. in female-headed households are females. By comparison, the vast majority of male heads of household are married. 6 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Figure 5.  Percent of Households with Members Younger Than 18, According to Sex of Head of Household Source: L2CU household survey (June/July 2018). Note: Figure 5 shows the percentage of households that have members younger than 18. Approximately 37 percent of households had at least one grandchild of the head of household living in the household. Figure 6.  Age of Head of Household Source: L2CU household survey (June/July 2018). Note: Figure 6 shows the age of female and male heads of household. The average age of female heads of household was 57.0, and the average age of male heads of household was 49.7. Nearly one-third of female heads of household were 70 or older. Household Survey Profile 7 Figure 7.  Household Consumption Quintiles, According to Sex of Head of Household Source: L2CU household survey (June/July 2018). Note: Figure 7 shows the consumption levels of female-headed households and male-headed households, by consumption quintiles, where Q1 = up to UZS 2.107 million; Q2 = UZS 2.108 million to UZS 2.960 million; Q3 = UZS 2.961 million to UZS 4.076 million; Q4 = UZS 4.077 million to UZS 6.169 million; and Q5 = more than UZS 6.169 million. A higher percentage of female-headed households are likely to be in the lowest quintile. “Female-headed households FHHs were less likely than MHHs to report that members of the household worked. FHHs had fewer noncontinuous were less likely than male- workers, including seasonal and informal workers, than MHHs. FHHs and MHHs had a similar number of children headed households to (household members younger than 15) and relied on report that members of the remittances in equal measure. household worked.” 8 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey 3. FINDINGS OF THE L2CU SURVEY “Across the country, the Across the country, the survey results provided little indication that FHHs had less access to energy services, survey results provided but they had much greater difficulty paying for utilities and basic needs. Households that were unable to pay for utilities little indication that female- reduced food consumption and borrowed money to pay for headed households had less basic needs. Critical findings were in four areas: access to (and quality of) energy services, affordability and ability to access to energy services, pay, coping mechanisms, and priorities for public services and infrastructure. but they had much greater difficulty paying for utilities Access to (and Quality of) Energy Services and basic needs” The L2CU data show that FHHs had slightly greater access to central heating, hot water, and natural gas than Outside Tashkent, few households had central heating MHHs (figure 8). There were no differences in access to (figure 9). A greater percentage of FHHs had access to electricity according to the sex of the head of household heating than of MHHs, especially in Bukhara region (figure 9). (figure 8, table 1). Figure 8.  Access to Energy Sources, According to Sex of Head of Household Source: L2CU household survey (June/July 2018). Note: Figure 8 shows the percentage of male-headed households and female-headed households that have access to four sources of energy: central heating, electricity, hot water, and natural gas. Female-headed households had slightly greater access to central heating, hot water, and natural gas than male-headed households. Findings of the L2CU Survey 9 Figure 9.  Percentage of Households with Central Heating, According to Region and Sex of Head of Household Source: L2CU household survey (June/July 2018). Note: Figure 9 shows the percentage of female-headed households and male-headed households in each region that have access to central heating. In Tashkent city, the majority of female-headed households and male-headed households have access to central heating. Respondents to the L2CU monthly panel surveys few differences in water supply interruptions between FHHs experienced few interruptions in electricity. Electricity and MHHs (figure A1). was available to most households with few disruptions in all regions and at all survey waves; disruptions in the supply There were important differences in water supply of electricity are rare and seldom last more than a few interruptions between regions. Of households that hours (figure 10). There was little difference between FHHs experienced water disruptions, those in Jizzakh, Republic of and MHHs in number of hours of access to electricity. This Karakalpakstan, and Khorezm experienced more frequent finding holds across the September 2018, December 2018, interruptions in water supply access (figure A2). and March 2019 survey rounds (table 1). Likewise, there were 10 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Figure 10.  Number of Hours Household Had Access to Electricity Day Before Survey Administered, According to Region Source: L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 10 shows the percentage of households that had electricity for zero to 24 hours the day before the survey. Disruptions in the supply of electricity are rare and seldom last more than a few hours. Table 1.  Number of Hours Household Had Access to Electricity Day Before Survey Administered, According to Survey Round and Sex of Head of Household Survey round Head of household 0 to 12 hours 13 to 23 hours 24 hours n (%) 1 Male 8 (0.6) 129 (10.4) 1,103 (89.0) (Sept 2018) Female 1 (0.4) 29 (11.0) 234 (88.6) 4 Male 8 (0.6) 344 (27.6) 895 (71.8) (Dec 2018) Female 2 (0.8) 61 (23.1) 201 (76.1) 7 Male 9 (0.7) 470 (37.6) 770 (61.6) (Mar 2019) Female 0 (0.0) 84 (31.6) 182 (68.4) Source: L2CU monthly panel survey (September 2018/December 2018/March 2019). There was little difference between female-headed households and male-headed households in the number of hours that they had access to electricity. Findings of the L2CU Survey 11 The L2CU data show that satisfaction levels with electricity provision were higher than for any other type of infrastructure or public service (figure 11). There were no differences in satisfaction between MHHs and FHHs or between male and female respondents. Figure 11.  Satisfaction with Public Service Provision, According to Sex of Head of Household and Respondent Source: L2CU household survey (June/July 2018). Note: Figure 11 shows the percentage of female-headed households and male-headed households that consider public infrastructure and energy provision to be “satisfactory” or “good.” The female-headed household (FHH) column includes female and male respondents in female-headed households. Data for male-headed households (MHH) is disaggregated by the sex of the respondent. There were no differences in satisfaction between male-headed households and female-headed households or between male and female respondents. 12 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Affordability and Ability to Pay for Utilities The L2CU data show that, on average, MHHs paid more for utilities than FHHs (figure 12).1 The difference was only statistically significant for electricity and natural gas. Figure 12.  Amount of Last Payment for Utilities (Logged), According to Sex of Head of Household Source: L2CU household survey (June/July 2018). Note: Figure 12 shows the amount of the last payment for utilities among female- and male-headed households. Although, on average, male- headed households pay more for utilities than female-headed households, this difference was not statistically significant for central heating or hot water. 1 There was no correlation between household size and amount households reported spending on utilities. Findings of the L2CU Survey 13 Figure 13.  Cost of Utilities as Percentage of Household Budget, According to Sex of Head of Household (Log Scale) Source: L2CU household survey (June/July 2018). Note: Figure 13 shows the percentage of the household budget in female- and male-headed households that was spent on utilities. Central heating and natural gas were, on average, the costliest utilities as a percentage of budget for female- and male-headed households. Electricity had the least amount of variability. FHHs and MHHs spent a similar percentage of their total The majority of households surveyed were able to pay for household budget on central heating and electricity utilities. Jizzakh, Syrdarya, and Tashkent city had the (figure 13). FHHs spent slightly more on hot water and highest percentage of households reporting that they natural gas than MHHs; these differences were statistically were unable to pay for utilities (figure 14). FHHs found it significant. Central heating and natural gas were, on harder than MHHs to pay for basic utilities (table 2). FHHs’ average, the costliest utilities as a percentage of budget inability to pay differed according to region, being highest for female- and male-headed households. Electricity had in Bukhara, Kashkadarya, Syrdarya, and Surkhandarya the least amount of variability. (figure 15). 14 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Figure 14.  Proportion of Households Unable to Pay for Utilities, According to Region (All Survey Rounds) Source: L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 14 shows the percentage of households in each region that were not able to pay for utilities. Most households were able to pay for utilities. “Female-headed households’ inability to pay differed according to region, being highest in Bukhara, Kashkadarya, Syrdarya, and Surkhandarya.” Findings of the L2CU Survey 15 Figure 15.  Inability to Pay for Utilities, According to Region and Sex of Head of Household (All Survey Rounds) Source: L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 15 shows the percentage of male- and female-headed households in each region that were not able to pay for utilities. There were four areas with at least 20 percent of households unable to pay for utilities: Jizzakh, Syrdarya, Tashkent region, and Tashkent city. Female-headed households reported higher rates of inability to pay in nearly every region. Table 2.  Percentage of Households Unable to Pay for Utilities in Past 30 Days, According to Sex of Head of Household Survey round Head of household Yes No Have not Chi-square test Chi-square received bill yet statistic p-value % 1 Male 18.6 78.6 2.7 22.0 <0.001 (Sept 2018) Female 31.0 63.6 5.3 22.0 <0.001 4 Male 13.9 85.3 0.8 27.8 <0.001 (Dec 2018) Female 27.3 72.0 0.7 27.8 <0.001 7 Male 14.1 83.9 2.0 25.4 <0.001 (Mar 2019) Female 27.1 71.8 1.1 25.4 <0.001 Source: L2CU monthly panel survey (September 2018/December 2018/March 2019). Female-headed households found it harder than male- headed households to pay for basic utilities. 16 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey The links between the sex of the head of household Tashkent region as the reference category).2 A measure of and ability to pay for utilities were investigated using household-level social capital, constructed as an additive regression analysis. The authors estimated the relationship index of five L2CU measures of different aspects of social between ability to pay for utilities and the following capital and community trust, is also included.3 High levels characteristics: sex and age of head of household, whether of social capital could enable households to borrow money at least one household member worked for pay, number of from family and friends in times of need. Standard errors household members working for pay divided by number of are clustered at the regional level. The authors report on household members, log of total household earnings divided predicted probabilities that are statistically significant at by number of household members, consumption quintile, the p<0.05 level and are calculated from the regression urban or rural residence, survey round, and region (using model in the analysis that is presented in this report. Figure 16.  Predicted Probability of Inability of Female-Headed Households in Tashkent Region with at Least One Member Working for Pay to Pay for Utilities Sources: L2CU household baseline survey (June/July 2018) and L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 16 shows the predicted inability of female-headed households in Tashkent region with at least one employed member to pay for utilities. A typical female-headed household in an urban area of Tashkent region was much more likely to report being unable to pay for utilities than a female-headed household in a rural area. 2 The authors estimated alternative regression models with different sets of control variables, including household size, whether a household member had lost a job in the last seven days, number of seasonal laborers per capita in household, and different sources of household income. The coefficient estimates from these alternative models were similar to those of the base model. 3 The survey module used to construct the social capital index asked respondents whether they agreed with the following statements: “In this community, it is safe to entrust your home to your neighbor when traveling,” “I can trust someone to take care of my children for a few days while I’m away,” “If you leave your bicycle or motorcycle outside unlocked, you are worried that someone will steal it,” In this village, or neighborhood, most people know each other,” “When neighbors have an accident, other community members are willing to help.” FHHs were significantly less likely to feel safe leaving their home or their children in the care of a neighbor (figure A3). Findings of the L2CU Survey 17 The regression results indicated that ability to pay for MHHs (figure 17). Poor FHHs in rural and peri-urban areas utilities varied according to income level (consumption), had greater difficulty paying for utilities than MHHs in the location (urban vs rural and peri-urban), and sex of the same areas (figure 17). FHHs in Tashkent and the Republic head of household. Poor FHHs in urban areas struggled of Karakalpakstan found it much harder to pay for utilities much more to pay for utilities than poor FHHs in peri-urban than FHHs in other regions, such as Navoi and Ferghana or rural locations (figure 16). Poor urban FHHs also had (figure 18). These results are consistent with the regression much greater difficulty paying for utilities than poor urban analysis of household ability to afford food (figure A4). Figure 17.  Predicted Probability of Inability to Pay for Utilities for Households in Tashkent Region with at Least One Member Working for Pay Sources: L2CU household baseline survey (June/July 2018) and L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 17 shows the predicted inability of households in Tashkent region that have at least one employed member to pay for utilities. Female- headed households are more likely to be unable to pay than male-headed households in rural and urban areas. 18 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Figure 18.  Predicted Probability of Inability to Pay for Utilities for Urban Female-Headed Households with at least One Member Working for Pay Sources: L2CU household baseline survey (June/July 2018) and L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 18 shows the predicted inability of female-headed households in urban areas that have at least one employed member to pay for utilities. The regional differences are generally smaller than the gender gap, but there are still significant differences. “For female-headed households, the difficulty of paying for utilities was constant at every level of social capital.” Findings of the L2CU Survey 19 For FHHs, the difficulty of paying for utilities was than rural and peri-urban FHHs in Ferghana valley. For constant at every level of social capital (figure 19). This MHHs, higher levels of social capital were associated with was true for FHHs living in Tashkent region, as well as greater ability to pay for utilities. This may indicate that the Ferghana valley (figure 20). There were nevertheless MHHs have tighter-knit social networks to which they can regional variations, as urban FHHs in Tashkent region were turn for financial assistance. much more likely to report being unable to pay for utilities Figure 19.  Predicted Probability of Being Unable to Pay for Utilities (Urban Household in Tashkent Region Lowest Consumption Quintile) Sources: L2CU household baseline survey (June/July 2018) and L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 19 shows the predicted inability of urban households in the lowest consumption quintile to pay for utilities. Greater social capital, measured according to answers to a series of questions about trust in one’s community, was associated with greater ability of male-headed households to pay for utilities. 20 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Figure 20.  Predicted Probability of Being Unable to Pay for Utilities (Female-Headed Households in Lowest Consumption Quintile Living in Tashkent Region vs. Ferghana Valley) Sources: L2CU household baseline survey (June/July 2018) and L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 20 shows the predicted inability of female-headed households in the lowest consumption quintile to pay for utilities. Rural and peri- urban female-headed households in Ferghana valley were much less likely to report being unable to pay for utilities than urban female-headed households in Tashkent region. “For male-headed households, higher levels of social capital were associated with greater ability to pay for utilities. This may indicate that male-headed households have tighter-knit social networks to which they can turn for financial assistance. ” Findings of the L2CU Survey 21 Coping Mechanisms The regression results suggest that poor FHHs in household members were also more likely than MHHs with urban areas were much more likely to reduce their food no employed household members to reduce their food consumption to pay for basic household needs than consumption to pay for basic household needs (figure 21). poor MHHs in urban areas. Poor FHHs with employed Figure 21.  Predicted Probability of Reducing Food Consumption to Pay for Basic Needs for Urban Households in Tashkent Region Sources: L2CU household baseline survey (June/July 2018) and L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 21 shows the predicted probability of urban households reducing their food consumption to pay for basic needs. Households in the lowest consumption quintile were much more likely to consume less food to pay for basic needs, yet male-headed households with employed members in the lowest quintile were only slightly more likely to reduce their consumption of food than female-headed households without employed members (e.g., households depending on benefits or remittances). 22 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey The regression results suggest that FHHs draw on social needs was 54 percent (vs. 38 percent for a poor MHH living capital to a much less extent than MHHs when faced with in the same location) at the highest levels of social capital having to reduce their food consumption to pay for basic and 66 percent (vs. 50 percent for a poor MHH living in the needs. For a poor FHH in Tashkent region, the predicted same location) at the lowest levels of social capital (figures probability of reducing food consumption to pay for basic 22 and 23). Figure 22.  Predicted Probability of Reducing Food Consumption to Pay for Basic Needs (Female-Headed Households in Lowest Consumption Quintile Sources: L2CU household baseline survey (June/July 2018) and L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 22 shows the predicted probability of female-headed households in the lowest consumption quintile reducing food consumption to pay for basic needs, according to their level of social capital. “Female-headed households draw on social capital to a much less extent than male-headed households when faced with having to reduce their food consumption to pay for basic needs. ” Findings of the L2CU Survey 23 Figure 23.  Predicted Probability of Reducing Food Consumption to Pay for Basic Needs (Male-Headed Households in Lowest Consumption Quintile) Sources: L2CU household baseline survey (June/July 2018) and L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 23 shows the predicted probability of male-headed households in the lowest consumption quintile reducing food consumption to pay for basic needs. Social capital benefits male-headed households more than female-headed households, yet those differences do not offset the large regional disparities between the mostly urban Tashkent region and the predominantly rural Ferghana valley. The L2CU monthly panel data show that a statistically The share of respondents that reported borrowing money significantly larger share of FHHs than MHHs borrowed to pay for basic needs varied across regions. In Navoi, money to pay for basic needs. In September 2018, 28.8 14 percent of FHHs and 19 percent of MHHs reported percent of FHHs borrowed money to pay for basic needs borrowing money; in Bukhara, 39 percent of FHHs and 30 (vs. 22.7 percent of MHHs) (table 3). In March 2019, this percent of MHHs reported doing so (figure A5). share declined to 18.8 percent for FHHs and 13.2 percent for MHHs. This, combined with FHHs’ higher utility bills and greater relative inability to pay their bills, indicates that FHHs found it harder to access loans. 24 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Table 3.  Households That Borrowed Money to Pay for Basic Needs, According to Sex of Head of Household Survey round Head of household Yes No Chi-square test Chi-square statistic p-value % 1 Male 22.7 77.3 4.05 0.04 (Sept 2018) Female 28.8 71.2 4.05 0.04 4 Male 16.8 83.2 3.06 0.08 (Dec 2018) Female 21.6 78.4 3.06 0.08 7 Male 13.2 86.8 5.17 0.02 (Mar 2019) Female 18.8 81.2 5.17 0.02 Source: L2CU monthly panel survey (September 2018/December 2018/March 2019). Table 3 shows that a larger share of female-headed households than male-headed households borrowed money to pay for basic needs. There was little regional variation in the percentage is much less common than borrowing money, 22 percent of households that borrowed money to pay for basic of households in Jizzakh and 23 percent in Kashkadarya needs (figure 24). The highest rates were in Jizzakh reported doing so (figure 25). and Kashkadarya regions where nearly 25 percent of households reported borrowing money for this purpose. Although selling household assets to pay for basic needs Figure 24.  Proportion of Households That Took Out Loans to Pay for Basic Needs, According to Region (All Survey Rounds) Source: L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 24 shows the percentage of households that reported taking out loans to pay for basic needs. There was little regional variation in the percentage of households that took out loans to pay for basic needs. The highest rates were in Jizzakh and Kashkadarya, where nearly 25 percent of households reported taking out loans. Findings of the L2CU Survey 25 Figure 25.  Proportion of Households That Sold Assets to Pay for Basic Needs, According to Region (All Survey Rounds) Source: L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 25 shows the percentage of households that reported selling assets to pay for basic needs. There was little variation in the percentage of households that sold assets to pay for basic needs. The highest rates were in Jizzakh and Kashkadarya, where nearly 10 percent of households reported selling assets. By far the most common coping mechanism to pay for basic to pay for basic needs in all but three regions (Surkhandarya, needs was taking money from savings. Of the households that Tashkent city, and Tashkent region) (figure 26). have savings accounts, the majority took money from savings Figure 26.  Proportion of Households That Took Money from Savings to Pay for Basic Needs, According to Region (All Survey Rounds) Source: L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure 26 shows the percentage of households that took money from savings accounts to pay for basic needs. In most regions, approximately half of households with savings had to use some of this money to pay for basic needs. 26 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey More FHHs than MHHs reported that their financial wellbeing to be worse or much worse over the following 30 wellbeing was worse than 30 days earlier (figure 27). days (figure 28). Similarly, more FHHs than MHHs expected their financial Figure 27.  Assessment of Financial Wellbeing vs. 30 Days Ago, According to Sex of Head of Household Source: L2CU monthly panel survey (September 2018/December 2018/March 2019). Note: Figure 27 shows the assessment that female- and male-headed households make of their financial wellbeing in the 30 days before the survey. Female-headed households are more likely than male-headed households to report that their financial wellbeing deteriorated over the previous 30 days. Figure 28.  Expectation of Financial Wellbeing Over the Next 30 days, According to Sex of Head of Household Source: L2CU monthly panel survey (September 2018/December 2018/March 2019). Note: Figure 28 shows the expectation that female- and male-headed households have of their financial wellbeing in the 30 days following the survey. Female-headed households are more likely than male-headed households to report that they expect their financial situation to worsen over the following 30 days. Findings of the L2CU Survey 27 Priorities for Public Services and Infrastructure No substantive differences in priorities were found between MHHs and FHHs or between male and female respondents. More citizens view piped gas, central heating, drinking water, and paved roads as priorities for government attention than other infrastructure and services (figure 29A). More jobs for men and women and better business opportunities were top priorities (figure 29B). Figure 29.  How High a Priority Is It to Improve (A) Public Services and Infrastructure, and (B) Policies, According to Sex of Head of Household and Respondent A 28 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey B Source: Household survey (June/July 2018). Note: Figure 29 A and B shows the priority ascribed to various public services, infrastructure, and policies by female- and male-headed households. Female-headed households and male-headed households did not differ meaningfully in their priorities for public service improvements. Overall, the most important priorities were jobs, opportunities for businesses, access to paved roads, and access to public utilities such as piped gas and water. The female-headed households (FHH) column includes female and male respondents. Data for male-headed households (MHH) is disaggregated by the sex of the respondent. Findings of the L2CU Survey 29 4. CONCLUSION Although there was little indication that FHHs had less access to energy services (electricity, heating, hot “Female-headed households water, natural gas) in Uzbekistan, they found it much were more likely than male- more difficult than MHHs to pay for utilities and basic needs. FHHs were more likely than MHHs to reduce food headed households to reduce consumption and borrow money to pay for basic needs. High levels of community-level social capital mitigated food consumption and the high cost of utility bills for poor MHHs but not FHHs. borrow money to pay for Additional analysis is needed to explain why FHHs faced difficulties meeting basic needs given that there are few basic needs.” differences in household consumption levels between FHHs and MHHs. For example, future research could address how the composition of FHHs—including number of household Tariff levels that do not reflect FHHs’ difficulties in dependents—affects their ability to meet basic needs. affording utilities will constrain their access to energy. Public consultation processes must be sensitive to the Measures to upgrade the energy infrastructure to unique needs of FHHs when assessing communities’ ability increase households’ energy efficiency and energy subsidy and willingness to pay. Connection or user fee requirements reforms require accompanying measures to protect poor that do not offer affordable options—such as revolving households and FHHs from high bills and energy upgrade funds, grants, and affordable credit facilities to increase costs. A variety of mechanisms and pricing options can be household connectivity—will not meet the needs of poor used to enable women to pay for greater energy access, households, particularly women. One policy option would be including energy efficiency improvements (World Bank to charge poor households and FHHs lower rates for initial 2015b), such as: usage and then increase rates as consumption increases. Loans or staggered payment structures may also increase • Financing mechanisms to enable poor households and access where initial start-up and hook-up costs are high. FHHs to pay connection fees or buy energy-efficient appliances (e.g., offer low to 0 percent interest financing, The findings also suggest that FHHs do not benefit utility on-bill financing,4 public grants, energy efficiency from community social capital to the same extent as credit lines through a development bank) (World Bank MHHs. FHHs may have fewer friends and relatives to Group 2018) turn to for financial support. Interventions to increase • Instituting pay-as-you-go models, smart meters, and FHHs’ connectivity to other community members and progressive tariffs5 access to financial, social, and other resources would help • Waiving import duties for renewable energy products increase their resilience. This may involve greater efforts to reduce costs on the part of neighborhood citizen assembly members to • Instituting alternative payment methods (e.g., cash provide outreach to FHHs and bring them together with or in-kind, enabling remote payment for the urban other community members in community-wide meetings husbands or relatives of rural dwellers) and activities, such as participatory planning to inform government infrastructure investments in villages or participatory budgeting.6 4 On-bill financing refers to a loan that a utility makes to a customer, the proceeds of which pay for energy efficiency improvements. Customers make regular monthly loan payments on their utility bill until the loan is repaid. 5 Pay-as-you-go models allow low-income customers to pay off energy efficiency investments in flexible payment terms, usually over long periods of time, while creating a credit history through mobile payments. Smart meters allow consumers to monitor energy use and help them identify activities they are spending the most on so that they can make changes to save energy and pay less on their bills. Progressive tariffs rise with increasing levels of electricity consumption; for instance, monthly consumption in excess of 400 kWh is charged at nearly double the average cost of supply. They reward low consumption and involve other measures such as individual metering and often include assistance (e.g., subsidized tariffs) for poorer households. 6 Neighborhood citizen assemblies are community self-governing bodies that perform functions such as distributing social welfare payments, community policing, and maintaining villages’ cleanliness. Although previously informal, they are now enshrined in Article 105 of Uzbekistan’s Constitution as territorial self-government organizations. Additional research is needed to understand the extent to which they already provide outreach to FHHs and how they can improve this outreach. A recent decree requires that at least 10 percent of district-level budgets be prepared based on citizen feedback (Presidential Decree of August 22, 2018, “About the measures to provide openness of budget data and active engagement of citizens in budget process”). 30 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey To reduce FHHs’ vulnerability and increase their resilience, the government of Uzbekistan could also strengthen its support to FHHs for business development “Tariff levels that do not and entrepreneurship. National programs, such as the Every Family Is an Entrepreneur Program, that provide reflect female-headed collateral-free loans to women, including FHHs, could complement the loans with mentoring support. The World households’ difficulties Bank-financed Ferghana Valley Enterprise Project is an in affording utilities will example of how the government could provide mentorship and business development support to entrepreneurs. By constrain their access connecting FHH loan beneficiaries with successful business entrepreneurs, the government could help to increase to energy.” FHHs’ social networks and the number of people to whom they can turn for advice, support, and even clients. Conclusion 31 APPENDIX A Figure A1.  Days with Water Supply Disruptions in Previous Month for Households That Experienced Water Disruptions, According to Sex of Head of Household: (A) Round 1, (B) Round 3 A. Monthly Panel Survey Round 1 (September 2018) B. Monthly Panel Survey Round 3 (November 2018) Source: L2CU monthly panel survey (September 2018 and November 2018). Note: Figure A1 shows the percentage of female- and male-headed households that had experienced disruptions in water supply in the 30 days preceding the survey, according to the number of days the disruption lasted. The figure includes households that reported having water disruptions at least once in the previous 30 days (~11 percent of households). Water supply disruptions affected nearly all households in the sample. In November 2018, male-headed households were significantly more likely to have no water supply than female-headed households. 32 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Figure A2.  Days with Water Supply Disruptions in Previous Month for Households that Experienced Water Disruptions, According to Region Source: L2CU monthly panel survey (November 2018). Note: Figure A2 shows the percentage of households in each region that had experienced disruptions in water supply in the 30 days preceding the survey, according to the number of days the disruption lasted. The figure includes households that reported having water disruptions at least once in the previous 30 days (~11 percent of households). Most disruptions in most regions lasted between 20 and 30 days. Appendix A 33 Figure A3.  Measures of Social Capital, According to Sex of Head of Household Strongly agree Neither agree nor disagree Disagree Somewhat agree Somewhat disagree Don’t know Source: L2CU household baseline survey (June/July 2018) Note: Figure A3 shows the percentage of female- and male-headed households that reported relying on the various measures of social capital. Male-headed households tend to have higher levels of social capital in almost all categories, but female-headed households are slightly more likely to feel safe leaving a bicycle or motorcycle outside. 34 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey Figure A4.  Predicted Probability of Being Unable to Afford Food for All Household Members for a Household in Tashkent Region with at Least One Employed Member Sources: L2CU household baseline survey (June/July 2018) and L2CU monthly panel survey (combined September 2018 to March 2019). Note: Figure A4 shows the predicted probability of households in the Tashkent region being unable to afford food, according to their consumption level. Female-headed households were more likely to be unable to afford food than male-headed households in rural and urban areas. The gender gap and the rural-urban gap were considerably larger than the difference between the lowest and highest quintiles of consumption. Appendix A 35 Figure A5.  Percentage of Households That Borrowed Money to Pay for Basic Needs, According to Sex of Head of Household and Region Percentage of households that took out loans to pay for basic needs in past 30 days Sources: L2CU monthly panel survey (September 2018). Note: Figure A5 shows the percentage of female- and male-headed households in each region that reported borrowing money to pay for basic needs. In almost all regions, female-headed households were more likely to take out loans to pay for basic needs. The difference was most pronounced in Tashkent city, where female-headed households were twice as likely as male-headed households to take out loans, and in Surkhandarya, where male-headed households were 2.5 times as likely as female-headed households to take out loans. 36 Energy Vulnerability in Female-Headed Households: Findings from the Listening to Citizens of Uzbekistan Survey REFERENCES Ajwad, Mohamed Ihsan, Ilhom Abdulloev, Robin Audy, Stefan Hut, Joost de Laat, Igor Kheyfets, Jennica Larrison, Zlatko Nikoloski, and Federico Torracchi. 2014. The Skills Road: Skills for Employability in Uzbekistan. Washington, DC: World Bank. Canpolat, Ezgi, and Sophia Georgieva. 2019. Energy Subsidy Reform Assessment Framework: Incidence of Price Subsidies on Households and Distributional Impact of Reform—Qualitative Methods. Washington, DC: World Bank. Dutta, Soma, Annemarije Kooijman, and Elizabeth Cecelski. 2017. Energy Access and Gender: Getting the Right Balance. Washington, DC: World Bank. Hiller, Bradley Todd, Evgeny Kochkin, Yadviga Viktorivna Semikolenova, and Robertus A. Swinkels. 2016. Social Impact of Centralized Energy Services’ Conditions in Uzbekistan: A Qualitative Assessment. Washington, DC: World Bank. Republic of Uzbekistan. 2004. Uzbekistan Health Examination Survey 2002. Calverton, MD: Analytical and Information Center, Uzbekistan State Department of Statistics, and ORC Macro. UNDP (United Nations Development Programme). 2016. Promoting Energy Efficiency in Public Buildings in Uzbekistan. New York: UNDP. http://www.undp.uz/en/projects/project.php?id=155. World Bank. ND. “Study: Listening to the Citizens of Uzbekistan.” Washington, DC: World Bank. http://www.worldbank. org/en/country/uzbekistan/brief/l2cu. World Bank. 2013. Energy Intensity Level of Primary Energy (MJ/$2011 PPP GDP). http://data.worldbank.org/indicator/ EG.EGY.PRIM.PP.KD/countries. World Bank. 2015a. Toward Gender-Informed Energy Subsidy Reforms: Findings from Qualitative Studies in Europe and Central Asia. Washington, DC: World Bank. World Bank. 2015b. Tools to Understand Social Issues in Energy Tariff and Subsidy Reforms in Europe and Central Asia. Washington, DC: World Bank. World Bank. 2016. World Bank Migration Factbook. Washington, DC: World Bank. World Bank. 2018. District Heating Energy Efficiency Project, Project Appraisal Document. Washington DC: World Bank. World Bank Group. 2018. The Philippines: Options for Financing Energy Efficiency in Public Buildings. Washington, DC: World Bank. WHO (World Health Organization). ND. “Mortality from Household Air Pollution.” https://www.who.int/gho/phe/indoor_air_ pollution/burden/en/. References 37