Methodological Guidelines on Assessing Household Disability- Related Costs and Their Implication for Participation August 2025 © 2025 The World Bank and UNICEF 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This work is a product of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, links/footnotes and other information shown 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. The citation of works authored by others does not mean the World Bank endorses the views expressed by those authors or the content of their works. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. 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. Attribution—Please cite the work as follows: “World Bank and UNICEF. 2025. Methodological Guidelines on Assessing Household Disability-Related Costs and Their Implication for Participation. © World Bank.” Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e mail: pubrights@worldbank.org. Acknowledgments These guidelines were developed by a team led by Charlotte McClain-Nhlapo (World Bank) and Alexandre Cote (UNICEF). Ludovico Carraro (Consultant, World Bank) was the lead author. Team members included (in alphabetical order) Andrey Tretyak (World Bank), Daniel Mont (Center for Inclusive Policy), Debbie Bong (World Bank), Deepti Samant Raja (World Bank), Elizabeth Acul (World Bank), Lena Morgon Banks (The London School of Hygiene and Tropical Medicine), Mónica Pinilla-Roncancio (University of Los Andes, Bogotá), and Vlad Grigoras (World Bank). These guidelines build on findings from studies in multiple countries by UNICEF and the Center for Inclusive Policy. In addition to the core team, a collaboration of colleagues and researchers participated in two workshops and online discussions to identify the methods and processes provided in this report. The team is very grateful for the contributions of Charles Knox-Vydmanov (UNICEF), Chloe Blackwell (Centre for Research in Social Policy, Loughborough University), Hanna Alasuutari (World Bank), Katherine Hill (Centre for Research in Social Policy, Loughborough University), Ketevan Melikadze (UNICEF), Maia Bagrationi (MAC Georgia), Meenakshi Balasubramanian (Center for Inclusive Policy), Pamela Smith (SODIS Peru), and Zachary Morris (Stonybrook University). The team is grateful for feedback from World Bank peer reviewers Jose Cuesta Leiva, Nahla Zeitoun, Noah Yarrow, and Sameera Al Tuwaijri which strengthened the guidelines. The team is grateful for the support and guidance of Gopal Mitra (UNICEF), Nikolas Myint (World Bank), and Robin Mearns (World Bank). This work was funded by the World Bank’s Inclusive Education Initiative and the Human Rights Trust Fund, and the Arts & Humanities Research Council and Foreign, Commonwealth & Development Office (PENDA project), and UNICEF. Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Executive Summary Understanding Disability-Related Costs for Policy Development Persons with disabilities and their families face substantial disability-related costs which may not be covered by social programs and policies. These costs may be direct (out of pocket expenses) or indirect (reduced education or employment opportunities for persons with disabilities and their caregivers) and can be financially draining, increase vulnerability and poverty as well as limit participation and inclusion. These costs are related to the person’s specific impairment and the physical and social environment. Direct disability-related costs include disability-specific expenses (those required only by persons with disabilities) such as assistive products, and expenses for everyday goods and services, which are often higher than average for persons with disabilities (for example, higher transportation or utility costs if the person with disability spends more time at home) or require a specific composition (such as a different food diet). Participation levels directly influence disability-related costs: when children with disabilities attend school or adults engage in work and social life, expenses such as transportation and device maintenance rise. In contrast, staying at home may reduce direct costs but may increase unpaid care needs and indirect costs for families. Understanding the size and composition of these costs helps in establishing support policies that enable persons with disabilities to participate equally in society. A study on disability-related costs can generate three important outputs to support this goal: ö A mapping of existing government programs that support persons with disabilities, showing their current population coverage and level of support. ö An assessment of the disadvantage level experienced by persons with disabilities by measuring the size and composition of disability-related costs, to what extent existing government programs cover the costs, and how they contribute to unmet needs and persistent gaps with the rest of the population. ö Identification of possible intervention areas, in order of priority, that can start addressing these costs. The interventions needed to support the inclusion of persons with disabilities can be complex. They rarely (if ever) involve a single program but are a comprehensive package combining cash and in-kind support, which are required to address diverse needs according to the degree of functional difficulties and type of disability. The guidelines in this report explain the different methods for assessing disability-related costs and how to use the findings to identify priority areas. The report does not recommend specific interventions, however. Deciding on the interventions and determining their delivery methods and budgets occurs later in the policy development process. The evidence generated from a study on disability-related costs provides the rationale for intervention and identifies the scope of potential programs. New evidence triggers inclusion policies, which require social and political sensitivity. Thus, using participatory studies can build knowledge, raise awareness, facilitate consensus building, and strengthen the capacity of certain groups and nongovernmental organizations to articulate their demands. Methodological guidelines on assessing household disability-related costs and their implication for participation 4 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Methods for Estimating Disability-Related Costs Four main methods for measuring disability-related costs have been identified. They relate to budget standard measures and reflect broader approaches found in the literature on household needs assessment. The methods differ in how they use information to develop budget standards, drawing on surveys and behavioral analysis, expert input, or consensus-based approaches. Determining the optimal method for assessing costs depends on each method’s strengths and weaknesses, the country context, and the issues of interest. Often, a combination of the methods is necessary. The four main methods for measuring disability-related costs are: ö Consumption patterns analysis (CPA) compares spending patterns between households with and without a member with disabilities using existing household survey data to identify the source of disability-related expenses such as higher health care costs. ö Standard of living (SOL) assumes that higher spending or income leads to higher living standards, and that persons with disabilities must spend more to reach the same living standards as others. The SOL method does not identify specific expenses, however, but it can be combined with CPA (and considered one method called consumption patterns analysis and standard of living method [CPA-SOL]) to show the main disability-related spending sources. The SOL method relies on existing household survey data collecting income and expenditure data, including the identification of persons with disabilities and nonmonetary indicators of living standards. ö Goods and services used (GSU) method directly asks persons with disabilities about their disability- related expenses and then compares that with other households. Those surveyed must estimate their own spending, assume what others spend, and then calculate a difference, which from their experience is always positive. GSU provides straightforward calculations for disability-specific items but becomes more complicated for common goods. The method estimates average disability-related expenses, which can be disaggregated by disability type and severity, and can use quantitative or qualitative methods. ö Goods and services required (GSR) measures what persons with disabilities need to reach a target living standard or level of participation, not just actual spending. It provides the most comprehensive cost assessment by involving experts or focus groups.1 In Organisation for Economic Co-operation and Development (OECD) countries, GSR uses an established basket of goods and services for the general population for comparison, or in household surveys asks people about unmet needs and their costs. However, in low and middle income countries (LMIC), fragmented markets for disability-specific items and a lack of information on products and services availability make GSR more difficult to apply. An exploratory GSR is a more effective approach to this method, because it uses experts and persons with disabilities to identify needed goods and services and then costs the basket using prevailing prices in the country. The different methods highlight the fundamental challenges of researching for persons with disabilities: quantitative methods struggle to obtain representative samples, and qualitative studies can be influenced by the imposition of common preferences2. Thus, combining methods is recommended most often. Persons with disabilities and their families face three key barriers that limit accurate cost measurement: lower incomes because of indirect costs such as reduced employment opportunities and caregiving related opportunity costs, unavailable disability-specific goods and services (such as assistive devices) because of underdeveloped markets, and people’s lack of information, and household discrimination leading to 1. Using a quantitative GSR approach, people are asked what they spend on disability-related items and whether what they consume meets their needs and if not, how much more they would need and spend. These questions assume people’s knowledge about disability-specific goods and services and rely on declared preferences to calculate the overall disability-related costs. 2. Assuming that every person with a certain type of disability has identical preferences. Methodological guidelines on assessing household disability-related costs and their implication for participation 5 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix unequal resource allocation, ignoring or repressing the needs of the person with disability. The four methods respond differently to those constraints and market limitations. CPA-SOL overcomes income constraints by calculating costs based on reaching a specific living standard, but it could fail to detect significant costs when disability-specific items are unavailable, unaffordable or unknown to people, or persons with disabilities face systemic exclusion, even though the needs exist. The GSU method measures expenses from income and living standards, providing biased estimates of the true costs subject to the income, availability, and discrimination constraints. GSR overcomes those constraints by considering required rather than actual spending, but it must carefully assess target participation levels and explain the experts’ assumptions. Advantages and Disadvantages of the Different Methods CPA-SOL uses existing data, as long as surveys include relevant questions and sufficient sample sizes to capture enough households with persons with disabilities. The method provides average disability-related expenses needed to equalize living standards, useful for correcting poverty measures and to inform design of cash transfers. It can also assess existing program coverage and produce some quantitative measures of unmet needs for health, education, and other services. However, CPA-SOL typically distinguishes only two or three subgroups by degree of functional limitations and not disability type, provides few details on expense composition, offers limited information on unmet disability needs for designing in-kind services, and allows minimal government and civil society engagement. GSU focuses exclusively on persons with disabilities, providing detailed information on disability-related expenses and exploring differences across disability types and functional difficulties. Unlike CPA-SOL, GSU can provide more details on expense composition, and it can inform both cash transfer and in-kind service design. It requires focused data collection, enabling greater ability to raise public awareness and opportunities to build political support. However, the method remains subject to income, availability, and discrimination constraints, producing biased estimates of true disability costs. Budget estimation proves challenging, because survey scope variations depend on available sampling frames. GSR stands as the only method that measures what persons with disabilities need to reach a target living standard or level of participation, not just actual spending. Like GSU, it also requires focused data collection. The exploratory GSR has a strong participatory approach and can inform comprehensive responses to support needs, especially in developing services. It focuses on different groups of persons with disability (age, location, impairments) and their support needs, disaggregating costs into policy categories and providing cost structure information instead of average estimates. However, GSR provides only illustrative cases, not representative estimates. All methods can allow mapping the existing programs (understanding and documenting how services are organized, such as what is available, which agency provides it, who is eligible, what is covered, and other relevant information). However, CPA-SOL and GSU could also assess the programs’ effectiveness in coverage of potentially eligible people and the level of support received. Selecting a method depends on the country context, data availability, and the study’s main objectives. No method provides all the answers, and thus combining approaches is suggested, where possible. Study Budgets for the Different Methods The approximate study budgets are $15,000 for the CPA-SOL (assuming use of already existing household survey data) and at least $50,000 for an exploratory GSR, but these figures differ depending on the type of analysis and how many groups of persons with disabilities are included. For the GSU method, budget Methodological guidelines on assessing household disability-related costs and their implication for participation 6 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix estimation is challenging because of variations in survey scope that depend on the available sampling frame. A phone survey is the least expensive option at $25,000, but the cost rises sharply if a population survey with disability screening is needed to build the sampling frame. Table ES.1: Advantages and Disadvantages of the Disability-Related Costs Measuring Methods Output CPA-SOL GSU (quantitative) Exploratory GSR Map of existing support Yes, also effectiveness Yes, also effectiveness Yes Assessment of disadvantage Quantification of costs Yes, but limited structure Biased, but structure Illustrative plus structure Correction of poverty measures Yes No No Diversity by disability types Very limited Yes Yes Identification of policy areas Cash support Yes, contributions Yes, contributions Yes, contributions In-kind Very limited Limited Yes Contributes to Political support Limited Yes Yes Population awareness Limited Yes Yes Resources and timeline Approximate budget $15,000 $25,000a $50,000 Required time 23 months 4–6 months 6 months Source: Original table for this publication. World Bank and UNICEF, 2025. Note: CPA = consumption pattern analysis; GSR = goods and services required; GSU = goods and services used; SOL = standard of living. a. Assuming it is possible to conduct a phone survey. Policy Implications and Developing a Support Package The study results can provide direct insights into possible support interventions, but they do not provide specific policy recommendations, which requires examining possible interventions and understanding their implementation and budget. Social protection interventions include cash support for general disability-related expenses, and in-kind provision of disability-specific goods and services such as assistive devices, habilitation and rehabilitation sessions, human assistance, and the like. Cash transfers can be designed to either address exclusively disability-related costs or provide general income support. However, when income support depends on a household income level assessment, the assessment must consider the role of disability-related costs. Estimates of average disability-related expenses provided by the SOL and GSU methods, and information on the structure of costs from the GSR, can provide various inputs to the design of disability allowances aimed at addressing part of the disability- related costs and guide the assessment of eligibility to poverty-targeted benefits. Methodological guidelines on assessing household disability-related costs and their implication for participation 7 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Additionally, information from GSU and GSR studies helps identify disability-specific items that can be supported through in-kind benefits (direct provision, subsidies, and concessions). In-kind benefits allow more flexibility in responding to large and diverse needs and allow the government to provide disability- specific goods and services, addressing market failures that limit their availability. Developing a response package to disability-related costs varies considering the degree of functional needs and cases in which high care support is needed. Package elements include access to health care, assistive products, habilitation and rehabilitation services, inclusive basic services, concessions to offset specific sources of disability costs, disability allowance, access to specific support services (for example, personal assistance and respite care), and access to income support. Two principles should guide response package development: ö Ensure equity of support across the country and for the diverse needs of persons with disabilities, and ö Begin by improving existing programs, develop new programs when needed, and then address coverage and level of support gradually. Equity of support should consider the diverse needs of people with different types of disabilities and degrees of functional limitation. Generally, cash transfers can address only some basic and generic needs (perhaps with some variation of value based on disability severity), but in-kind services must address the complexity and heterogeneity of needs. Development should always start with existing programs. Based on study findings, policy makers can identify ways to improve these programs by ensuring equitable implementation, then consider new interventions, starting with needs that cause the greatest inequities across groups of persons with disabilities. The possible interventions are many, so good practice is to prioritize them for sequenced implementation. Methodological guidelines on assessing household disability-related costs and their implication for participation 8 CONTENTS Acknowledgments 3 Executive Summary 4 Introduction 12 1.1 Why Assess Disability-Related Costs 13 1.2 Uses for a Study on Disability-Related Costs 14 1.3 Understanding Disability-Related Costs at the Household Level 15 Consumption Patterns Analysis 20 2.1 Implementing CPA 21 2.2 Limitations and Recommendations for CPA 23 Standard of Living Method 25 3.1 Choice of Standard of Living Indicator 29 3.2 Consumption or Income 30 3.3 Functional Form of Consumption Expenditure 31 3.4 Implementing the SOL Method 34 3.5 Limitations and Recommendations for the SOL Method 36 Goods andServices Used 37 4.1 Implementing the GSU Method 40 4.2 Limitations and Recommendations for the GSU Method 42 Goods and Services Required 43 5.1 Quantitative GSR 44 5.2 Comparative Qualitative GSR 45 5.3 Exploratory Qualitative GSR 47 5.4 Limitations and Recommendations for the GSR Method 53 Measures to Address Disability-Related Costs: Developing a Support Package 54 6.1 Cash Transfers to Address Disability-Related Costs 56 6.2 In-Kind Benefits 60 6.3 Designing the Package: Pros and Cons of Different Instruments 63 6.4 Developing a Response Package in Different Contexts 66 References 71 Appendix A. Methodologies Used to Estimate Disability-Related Costs 75 Tables Table ES.1: Advantages and Disadvantages of the Disability-Related Costs Measuring Methods  7 Table 1.1: Advantages and Disadvantages of the Disability-Related Costs Measuring Methods  19 Table 2.1: Consumption Patterns for Households with Children by Disability in the Philippines 22 Table 6.1: Eligibility Rules to Poverty Targeted Benefits Accounting for Disability-Related Expenditure 58 Table 6.2: Cash Benefits and Their Role in Addressing Disability-Related Costs 59 Table 6.3: Equity, Agency, and Implementation Considerations in Key Interventions 64 Table 6.4: Limited Package: Current Programs, Improvements, and Medium-Term Priorities 67 Table 6.5: Advanced Package: Current Programs, Improvements, and Medium-Term Priorities 69 Table A.1: Characteristics of the Different Methods on Key Dimensions 80 Table A.2: Differences in Outputs of Methodologies Used to Assess Disability-Related Costs 83 Table A.3: Examples of Research Questions Answered by the Different Methods 84 Figures Figure 1.1: The Role of Household-Borne Disability-Related Costs in Affecting Participation 13 Figure 3.1: Standard of Living and Disability-Related Costs 26 Figure 3.2: Consumption Expenditure by Living Standard and Disability Level for Two-Parent, Two-Child Households in the Philippines 27 Figure 3.3: Logarithmic Functional Form and Size of Disability-Related Costs 31 Figure 5.1: Use of Group Discussions to Determine Consensual Budget Standards 46 Figure 5.2: Exploratory GSR Implementation Overview 48 Figure 6.1: The main categories of social protection instruments, types of schemes and related functions across the lifecycle 55 Figure 6.2: Package of Support to Persons with Disabilities Depending on the Level of Recognized Needs and Disability Costs 63 Figure A.1: Different Budget Methods to Determine Needs 77 Boxes Box 1.1: Targeted Subsidies and In-Kind Transfers 13 Box 1.2: Inclusion and Exclusion of Subsidies’ Imputations 16 Box 2.1: Measuring Unmet Needs Using Common Questions in Household Surveys 23 Box 3.1: Correcting Poverty Measures for Incurred Disability-Related Costs 32 Box 3.2: Possible Disability Proxies and How to Measure Disability Severity 33 Box 3.3: Time Use Modules and Disability 34 Box 4.1: Surveys of Persons with Disabilities in the 1970s and 1980s in the UK 38 Box 4.2: Assessing the Effectiveness of Current Interventions for Persons with Disabilities 39 Box 4.3: Probability Sampling and Persons with Disabilities 41 Box 5.1: Example Definitions of Participation and Income Required to Achieve It 50 Box 5.2: Fictional Examples of the Type of Outputs from Illustrative Cases 51 Box 6.1: Adjusting Proxy Means Tests to Account for the Needs of Persons with Disabilities 59 Box 6.2: Designing the Structure of Concessions 61 Box 6.3: Developing a Response Package in Georgia 70 Box A.1: Identifying Persons with Disabilities in Household Surveys and Washington Group Questions 81 Abbreviations CPA consumption patterns analysis CPA-SOL consumption patterns analysis and standard of living method FGD focus group discussion GSR goods and services required GSU goods and services used LMIC low and middle income countries OPCS Office of Population Censuses and Surveys OECD Organization for Economic Cooperation and Development SOL standard of living All dollar amounts are US dollars unless otherwise indicated. 01. Introduction KEY MESSAGES Understanding the structure and burden of disability-related costs is essential to designing effective, inclusive social policies. A focused study can provide this foundation by mapping existing support programs, assessing the disadvantage these costs create, and identifying potential interventions. Such analysis does not produce policy recommendations directly but can informs later stages of program design and budgeting. Addressing these costs is both a matter of economic rationale and social equity. However, doing so requires better data and confronting broader political and institutional challenges. Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 1.1 Why Assess Disability-Related Costs Persons with disabilities and their families incur many costs for disability-related goods and services that arise because of a specific functional limitation and existing environmental barriers. For example, a person who uses a wheelchair (functional limitation) to move about cannot use public transportation if it is not wheelchair accessible (environmental barrier) and needs to use other, more expensive private transportation. Such costs that households incur create an obstacle to participating equally in society (figure 1.1). Figure 1.1: The Role of Household-Borne Disability-Related Costs in Affecting Participation Environmental barriers Household-borne disability-related Disability costs (lack of participation) Impairment and other personal characteristics Source: Original figure for this publication; World Bank and UNICEF 2025. Assessing the size and structure of disability-related costs and their possible impact on the exclusion of persons with disabilities and their families builds the rationale for public policy intervention and identifying areas for program development aimed at reducing these costs and increasing social participation. Public policy intervention can absorb some household costs through in-kind transfers and targeted subsidies (discounts, concessions,3 fee waivers, or exemptions; see box 1.1 for more details and examples), through cash transfers to households or by reducing environmental barriers. When the state absorbs some of these costs, the costs remain, but who pays for them and the method of payment change. In-kind transfers, targeted subsidies, and cash transfers provide support at the individual level. By contrast, reducing environmental barriers can eliminate the source of the cost. Environmental barriers include the physical infrastructure (such as inaccessible buildings), communication barriers (such as lack of captioning in video announcements), and discriminatory and biased social norms and attitudes. Box 1.1:   Targeted Subsidies and In-Kind Transfers Subsidies are in-kind support to reduce the cost of goods and services and can be targeted for some people for specific goods and services (concessions, exemptions, and fee waivers). Unlike cash transfers, which can be spent freely, subsidies are tied to a good or service. Subsidies for persons with disabilities promote fairness and address market failures. Some markets might not develop without government intervention, and thus the positive externalities associated with its consumption are lost. For example, assistive devices clearly benefit users but also reduce care and support needs and increase economic participation, thus benefiting society. Examples of subsidies for persons with disabilities include: transportation discounts for persons with disabilities; utilities discounts; partial or full fee waivers for certain health services or education; value added tax and other tax exemptions, such as on assistive devices; reduced rents for government housing; free social care and support services; and vouchers for personal assistance services for persons with disabilities. Source: World Bank and UNICEF, 2025. 3. Concessions in this context refers to price reductions based on eligibility, or price structures tied to social policy rather than market pricing. They can include administrative mechanisms, such as vouchers or preapproval. Methodological guidelines on assessing household disability-related costs and their implication for participation 13 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix The guidelines focus on assessing households’ direct costs and to what extent existing policies cover those costs. The assessment considers the current country context while comparing conditions across different areas. 1.2 Uses for a Study on Disability-Related Costs A study on disability-related costs can provide three main outputs: ö A mapping of existing programs and services that reduce disability-related costs and social protection programs providing direct cash support, which could include an assessment of program effectiveness such as accessibility and use of existing programs and services (population coverage and level of transfers). ö An assessment of the current disadvantage level of persons with disabilities by measuring the size and structure of disability related costs and the presence and size of existing gaps in key outcome indicators (such as enrollment rates and employment participation) ö Possible policy intervention areas that could reduce disability-related costs. The guidelines in this report cover the process of measuring household disability-related costs. They do not calculate the costs of a state intervention (for example, providing assistive technologies through the school system) and do not propose an intervention package. Developing policy measures or an intervention package to reduce these costs, along with implementation systems and budgets for interventions, can be done in three phases: understanding disability-related costs (clearly show the need and rationale for intervention), scoping different intervention packages (determine intervention priority and their possible sequence), and developing a budget and a delivery system (which is more demanding for new interventions). These guidelines focus primarily on the first phase but also mention the second phase. Methodological guidelines on assessing household disability-related 14 costs and their implication for participation Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 1.3 Understanding Disability-Related Costs at the Household Level Disability-related costs are the expenses that persons with disabilities and their families should incur and must cover to participate in society. The costs arise because of their impairment, functional limitations and existing environmental barriers. These costs may be direct (out of pocket expenses) or indirect (reduced education employment opportunities for persons with disabilities but also their caregivers). The level of participation sought by persons with disabilities also impacts disability related costs: if a child with disabilities goes to school it implies transport costs, possibly greater wear and tear of assistive devices, but if a child stays at home out of school, then there is more need for unpaid care and therefore higher indirect costs. Similarly, an adult engaging in economic and social activities will face higher disability costs than if they stay inactive at home. Disability-related costs can be disaggregated across different dimensions: ö Who bears the cost: the household or the state, and if the state, understanding the value of the subsidized items is necessary (whether provided free of charge or through discounts and concessions). ö Disability-specific and common items: whether only persons with disabilities need them (for example, assistive technology such as hearing aids or a wheelchair, braille books, and sign language interpretation) or they are common items that persons with disabilities need in larger amounts or in different ways than others. For example, people who are less mobile would require more home heating because they spend more time at home; or public transportation that is not accessible means people would need to use different and possibly more expensive transportation. ö Goods and services: a physical or tangible object, such as an assistive device or diapers (goods) and receiving a treatment or gaining a certain skill, such as fitting and adjustments of an assistive device or a speech therapy session (services). Providing goods versus services generally entails substantially different arrangements. ö Frequency of purchases: recurrent expenses (for example, monthly) and other less frequent expenses for buying durable items or one-off expenses. Indirect costs are borne by the household because persons with disabilities often have lower employment rates and lower educational attainment (and thus lower-paid jobs), or because family members who provide care need to give up job opportunities or accept lower-paid work to accommodate their caregiving responsibilities. Therefore, indirect costs imply lower household incomes. The lower employment and productivity are also public indirect costs that have a negative effect on the whole economy. High direct costs have an impact on indirect costs whenever they reduce employment participation or result in lower educational achievement. If they increase the dependency of the person with disability, more caregiving time is usually required, creating opportunity costs for family members. Therefore, reducing disability-related household direct costs increases the participation of persons with disabilities and reduces indirect costs for the household and at the public level, which is a strong economic rationale for intervention. The term “indirect costs” describes reduced income because of disability, but it still has a direct effect on a household’s well-being. The terminology helps explain how costs arise, but both lower income and additional expenses reduce what households with a person with disability can consume compared with similar households without a person with disability. Methodological guidelines on assessing household disability-related costs and their implication for participation 15 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix The term “indirect costs” describes reduced income because of disability, but it still has a direct effect on a household’s well-being. The terminology helps explain how costs arise, but both lower income and additional expenses reduce what households with a person with disability can consume compared with similar households without a person with disability. The guidelines in this report focus on direct costs but consider the measurement of indirect costs. State subsidies are considered only if they support private goods for specific persons and households. General subsidies (such as reduced electricity costs for all) or public goods (such as roads or playgrounds) are excluded. Subsidies that are available universally (such as free primary education for all children) are also excluded, even if the benefit differs by disability status (see box 1.2 for examples of when to estimate state subsidies). Box 1.2:   Inclusion and Exclusion of Subsidies’ Imputations Subsidies are in-kind support to reduce the cost of goods and services and can be targeted for some people for specific goods The following are examples of the type of subsidies that should be imputed or disregarded when assessing direct household disability-related costs. ö An assistive device is provided for free or at reduced cost: the full value of the cost should be included (what was paid plus the subsidy). ö A rehabilitation or other health service paid with a set discount (for example, 20 percent): accounting for the payment and the subsidy is necessary. ö All costs incurred to have access to services such as health and education (for example, transportation): they should be included. ö Education provided free of charge for all children, and the state pays for all services within the school: no need to account for the implicit subsidy. ö Children with disabilities have access to education accommodations and support but need to pay extra fees (for teacher assistants or for medical exams): these costs must be included. ö Children with disabilities are entitled to a set amount of day care per week in community centers for free: the subsidy needs to be included. ö The state adapted footpaths to be wheelchair accessible, and public announcements are barrier free: their implicit costs should not be included because they are public goods. Methods for Estimating Disability-Related Costs Four main methods for measuring direct disability-related costs have been identified. They relate to budget standard measures and reflect broader approaches found in the literature on household needs assessment. The methods differ in how they use information to develop budget standards, drawing on surveys and behavioral analysis, expert input, or consensus-based approaches. Determining the optimal method for assessing costs depends on each method’s strengths and weaknesses, the country context, and the issues of interest. Often, a combination of the methods is necessary. The four main methods for measuring disability-related costs are: ö Consumption patterns analysis (CPA): CPA compares the consumption expenditure patterns of households with and without a member with disabilities. Analyzing spending differences across consumption sub-aggregates (food, clothing, housing, utilities, health, transportation, communication, education, and so on), especially among households with similar overall spending, can reveal divergent priorities and thus the source of disability-related expenses, such as higher health care expenses. CPA uses existing household survey data with information on spending and with proxies to identify persons with disabilities. ö Standard of living (SOL): The SOL method assumes that higher spending or income leads to higher living standards. It also assumes that because of disability-related expenses and unmet needs, persons Methodological guidelines on assessing household disability-related 16 costs and their implication for participation Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix with disabilities must spend more to reach the same living standards as others. Therefore, the method estimates disability-related costs as the average additional expenditure incurred by households with members with disabilities to reach the same standard of living of other households. Costs can vary by living standard, the degree of functional difficulties, and the household composition. Although the SOL method calculates disability-related spending without identifying specific expenses, it can be combined with CPA—and considered as one method called consumption patterns analysis and standard of living method (CPA-SOL)—to show the main disability-related spending sources, because implementing the SOL approach relies on existing household survey data. However, the household survey must include nonmonetary indicators of living standards. ö Goods and services used (GSU): This method directly asks persons with disabilities to report their disability-related expenses for all goods and services used and then compares that with other households. People interviewed must estimate their own spending and assume what others spend and calculate a difference, which from their experience is always positive. The calculation can be straightforward for disability-specific items but more complicated for common goods. The method estimates average disability-related expenses, which can be disaggregated by groups of persons with disabilities and severity of functional limitations. The GSU approach can use either quantitative or qualitative methods. ö Goods and services required (GSR): This method looks beyond actual spending and use to measure what persons with disabilities need to reach a target living standard or level of participation. GSR provides the most comprehensive assessment of disability-related costs and not just expenditure, achieved with the involvement of experts or through focus group discussions.4 In Organisation for Economic Co-operation and Development (OECD) countries, this approach uses an already established basket of goods and services for the general population for comparison, or in household surveys asks people about unmet needs and their costs. However, these approaches are not easy to apply in LMIC because of a lack of information on the availability of services and products, and fragmented markets for disability-specific items. A more effective approach to this method, called an exploratory GSR, uses experts and persons with disabilities to identify the goods and services needed and then cost the basket based on prevailing prices in the country. The different methods highlight the fundamental challenges of researching for persons with disabilities: quantitative methods struggle to obtain representative samples, and qualitative studies can be influenced by the imposition of the same preferences across people with a certain type of disability. Thus, combining methods is recommended most often. Persons with disabilities and their families face significant barriers that limit how well disability-related costs can be measured. First, they often have lower incomes because of indirect costs such as reduced employment opportunities and caregiving responsibilities that affect household earnings, thus making it difficult to meet specific needs. Second, many disability-specific goods and services, such as assistive devices, might be unavailable because of underdeveloped markets and people’s lack of information about such goods and services. Third, discrimination within households can lead to unequal resource allocation, with the needs of the person with disability ignored or repressed. The four methods respond differently to those constraints and market limitations. The CPA-SOL method overcomes the income constraint by calculating disability-related costs based on reaching a specific living standard, and the living standard measure can be sensitive to unmet needs. However, if disability-specific items are unavailable in the country or unknown to people, or persons with disabilities are systematically excluded from participation, the method could fail to detect significant disability-related costs—not because the needs do not exist but because they go unrecognized and remain unmet. The GSU method tends to measure disability-related expenses from income and living standards and is likely to provide a 4. Using a quantitative GSR approach, people are asked what they spend on disability-related items and whether what they consume meets their needs and if not, how much more they would need and spend. These questions assume people’s knowledge about disability-specific goods and services and rely on declared preferences to calculate the overall disability-related costs. Methodological guidelines on assessing household disability-related costs and their implication for participation 17 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix biased estimate of the true costs of disability, and their measures are subject to the income, availability, and discrimination constraints. The GSR method overcomes those constraints by considering what is required, but it must carefully assess the target level of participation and explain the experts’ assumptions. Advantages and Disadvantages of Methods The CPA-SOL method uses existing survey data, provided the survey includes relevant questions and the sample size is large enough to capture households with persons with disabilities. It estimates average disability-related expenses needed to equalize living standards with other households, which can help adjust poverty measures. However, it typically distinguishes only two or three subgroups by degree of functional limitations, not by disability type. It provides limited detail on the composition of expenses and cannot disaggregate spending by type of disability. If relevant questions are included, the method can assess coverage of support programs and generate basic quantitative estimates of unmet needs in health, education, and other services. CPA-SOL results can suggest policy implications, particularly for recognizing disability-related expenses and designing cash transfers, though they offer limited support for planning in- kind services. Because it uses existing data, opportunities for government engagement, political support, and public awareness are more limited than with other methods. The method is relatively low-cost, with an estimated budget of $15,000 when based on existing household survey data. The GSU method focuses exclusively on persons with disabilities, enabling detailed analysis of disability- related expenses by type and degree of functional difficulty. It can also assess program coverage and support levels for potentially eligible individuals. Compared with CPA-SOL, GSU offers more detail on expenditure structure and can inform both cash and in-kind policy interventions. Like exploratory GSR, GSU requires dedicated data collection. Cost varies depending on the sampling approach: using a phone survey is the least expensive option at about $25,000, but a full population survey with disability screening, if needed to build the sampling frame, increases the budget significantly. The exploratory GSR method also focuses on disability types and support needs, disaggregating costs into policy categories. It does not estimate average costs but instead provides detailed information on cost structures and the specific goods and services needed for different impairments. It is the only method that estimates required—not just actual—expenditures to ensure equal participation. However, its results are illustrative, not representative. The method relies on a strong participatory approach and supports broad policy planning, especially for developing services. Like GSU, GSR allows more engagement with stakeholders and greater potential to build political support and public awareness. An exploratory GSR costs at least $50,000, depending on the number of disability types included and the analysis scope. All methods allow mapping existing programs (understanding and documenting how services are organized, such as what is available, which agency provides it, who is eligible, what is covered, and other relevant information). However, only CPA-SOL and GSU can assess how well programs reach potentially eligible people and the level of support provided. Table 1.1 summarizes the different methods’ advantages and disadvantages, focusing on their outputs and their potential to build political support and raise public awareness. Selecting a method depends on the country context, data availability and the study’s main objectives. No method provides all answers, and where possible, combining approaches is suggested. Methodological guidelines on assessing household disability-related 18 costs and their implication for participation Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Table 1.1: Advantages and Disadvantages of the Disability-Related Costs Measuring Methods Output CPA-SOL GSU (quantitative) Exploratory GSR Map of existing support Yes, also effectiveness Yes, also effectiveness Yes Assessment of disadvantage Quantification of costs Yes, but limited structure Biased, but structure Illustrative plus structure Correction of poverty Yes No No measures Diversity by disability types Very limited Yes Yes Identification of policy areas Cash support Yes, contributions Yes, contributions Yes, contributions In-kind Very limited Limited Yes Contributes to Political support Limited Yes Yes Population awareness Limited Yes Yes Resources and timeline Approximate budget $15,000 $25,000a $50,000 Required time 23 months 4–6 months 6 months Source: Original table for this publication; World Bank and UNICEF 2025. Note: CPA = consumption pattern analysis; GSR = goods and services required; GSU = goods and services used; SOL = standard of living. a. Assuming it is possible to conduct a phone survey. Appendix A is an overview of the existing methods for estimating disability-related costs. It highlights their main differences and complementarities, discussing their outputs and how to use them for policy development. The appendix gives readers enough information on the methodology that best suits their circumstances to understand the in-depth instructions on these methods in the next sections. In most cases, however, the best approach will require a combination of methods, also discussed in the appendix. The next sections cover the four approaches, followed by the final section, which discusses how to use the outputs of disability-related costs assessments—regardless of the specific method adopted in the research—to support the development of inclusive policies and intervention packages for persons with disabilities, which are the second and third stages of policy development. Two companion reports provide operational guidance for implementing the standard of living approach and analyzing quantitative surveys.5 5. For additional operational guidance on the SOL and exploratory GSR approaches, please visit: https://www.worldbank.org/en/topic/ disability Methodological guidelines on assessing household disability-related costs and their implication for participation 19 02. Consumption Patterns Analysis KEY MESSAGES Consumption patterns analysis (CPA) evaluates how households with and without a member with disabilities spend their money (including in-kind sources). Consumption expenditure is disaggregated into its subcomponents— such as food, clothes, utilities, education, and health—to highlight any differences in spending priorities. For example, CPA often shows that households with a member with disabilities tend to spend a larger share of their budget on health. For a more accurate comparison (if enough observations are available), the analysis focuses on households with similar size, family structure, and overall spending levels. However, the focus is not intended to meet a certain normative standard or basic needs threshold. A normative element can be introduced using the costs of basic needs methodology (Ravallion 1998). The analysis includes only households that spend enough for food to meet minimum nutritional requirements (adequate calories, macronutrients, vitamins, and minerals) and infers that their nonfood spending also reflects necessity. Specifically for determining the presence and size of disability- related costs, it is possible to compare households with and without members with disabilities separately to estimate any differences in the overall expenditure for households with members with disabilities. If a household’s food expenditure meets basic nutritional requirements, the nonfood share of households with members with disabilities is expected to be higher than that of other households, thus producing a separate and higher poverty line for such households. The difference in the value of the poverty line for households with and without poverty line provides an estimate of disability-related costs at very basic levels and can be used to calculate a specific equivalence scale for persons with disabilities. However, this approach requires a large household survey sample with enough observations of households with a member with disabilities. Although the analysis of consumption expenditure data for interhousehold welfare comparisons has limitations, comparing costs functions of households with and without a member with disabilities can provide lower-bound estimates of disability costs (see, for example, Jones and O’Donnell [1995]). This comparison is done by analyzing Engel curves and estimating specific equivalence scales for persons with disabilities: equivalence scales express the average needs of a person with certain demographic characteristics compared with other household members, or of a specific household composition compared with another. Using consumption expenditure data to determine equivalence scales faces an identification problem. Observed consumption behavior is affected by a household’s living conditions and the choice of household size, which is endogenous and thus makes it difficult to determine how household composition affects consumption expenditure (Pollak and Wales 1979). Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 2.1 Implementing CPA The CPA’s data source is household surveys that must include detailed information on household consumption expenditure and a breakdown of its components,6 along with information to identify persons with disabilities and provide some indication of their disability severity. National statistical offices usually conduct these surveys regularly and use them to monitor monetary poverty and living conditions. However, how well suited these surveys are for this analysis depends on what information they collected. Some lack the information needed to identify persons with disabilities (or a relevant proxy) and thus cannot be used to conduct the CPA. However, a survey should contain at least some information on possible proxies for disability, but the quality of questions—especially on functional difficulties—can vary significantly. Also, the questionnaire design might not capture other aspects relevant for this type of analysis, such as the possibility to impute subsidies received by persons with disabilities; and the level of detail on disability-specific expenditure (such as for assistive devices, rehabilitation, and personal assistance). When persons with disabilities in a country are entitled to subsidies (discounts or concessions) on specific expenditure items such as utilities, health, or transportation, the questionnaire should be able to identify whether they have used this support, so that the subsidy can be imputed into the full consumption expenditure. However, if at national level relatively few people are receiving subsidies, the survey might not include the information needed to impute subsidies, a clear limitation affecting the analysis. In these cases, its implication should be assessed relative to each country’s specific circumstances. Consumption expenditure should be comprehensive, but questionnaire design always involves trade-offs between the level of detail and the relative importance of the information collected. Therefore, many surveys do not collect information on some disability-specific expenditures, which does not necessarily mean that these expenditures are not collected. They could be grouped into various residual items. For example, a survey might not ask specifically about disability-related rehabilitation services (such as speech therapies, sessions with occupational therapists, or physiotherapies). Instead, it might include a question about “other services” within a broader section on medical services, which might not an effective way to ask about expenses if the recall period is inadequate or if recall bias occurs because respondents are not specifically reminded of these costs. However, respondents will likely report any significant expenditure that the questionnaire does not cover directly under “other services” questions in the survey’s different sections. Because of the Sustainable Development Goals’ emphasis on disability, more surveys now collect information on functional difficulties using the Washington Group set of questions and other proxies that help identify persons with disabilities. Therefore, the possibility of using existing surveys for CPA is higher now than it was 10 years ago, though improvement is still needed.7 This includes \ the need for better data on \ children with disabilities using the Child Functioning Module.8 Household surveys are recommended for this type of analysis, especially when they lead to engagement with the national statistical office to help improve questionnaire design. For example, some countries have replaced questions about official disability status with the Washington Group set of questions on functional difficulties. However, this is ineffective because it removes the possibility of analyzing the overlap between functional limitations and official disability status, or identifying cases in which official disability status is not associated with functional difficulties. Additionally, a lack of information on who holds official disability status prevents accurately identifying and imputing any subsidies provided to them. Ideally, questions on functional difficulties should be followed by a question on official disability status and its classification. 6. The United Nations Classification of Individual Consumption by Purpose provides a common breakdown of consumption expenditure: food, tobacco, and alcoholic beverages; clothing and footwear; housing and utilities; furnishing and equipment; health; transportation; communication; recreation and culture; education; restaurants and hotels; and other miscellaneous goods. 7. See evidence gathered by the Disability Data Initiative at https://www.disabilitydatainitiative.org/ (visited June 26, 2025). 8. See Washington Group website: https:/ /www.washingtongroup-disability.com/question-sets/wg-unicef-child-functioning-module-cfm/ (visited June 26, 2025). Methodological guidelines on assessing household disability-related 21 costs and their implication for participation Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Recent examples of CPA are included in a study conducted by Indecon International Research Economists for the government of Ireland Department of Social Protection, The Cost of Disability in Ireland, and a study published by UNICEF Philippines, Cost of Raising Children with Disabilities in the Philippines. In both studies, the CPA was only one component of the broader study of disability-related costs. The study in Ireland used household budget survey data collected by the Central Statistics Office and focused on households within 10 percent of the median income of households with a person with disability, thus restricting the analysis to households with a similar overall budget, allowing the study to identify any divergent patterns in the expenditure breakdown. The study in the Philippines used an ad hoc survey to collect the data, focusing on households with children and comparing the consumption expenditure patterns of households with or without a child with disability (with functional difficulties and a disability card). Such households have similar average overall consumption expenditure, but the households’ priorities for spending their budgets clearly differ. This analysis also included the state-provided subsidy through the disability identification card and information on how each household uses it. Table 2.1 summarizes the results of the study in the Philippines, where on average health expenditure for households with children with disabilities is three times as high (in percentage) that of other households. Table 2.1: Consumption Patterns for Households with Children by Disability in the Philippines Children with functional difficulties and disability card Consumption group Yes No Food 56.1 62.2 Alcohol and tobacco 1.5 2.2 Education 5.0 4.3 Health 11.6 3.7 Clothes 1.3 1.3 Utilities 9.7 10.9 Transportation 4.1 4.7 Communication 3.9 4.1 Personal care 3.3 3.4 Maintenance 2.0 2.1 Recreation 1.6 1.0 Total 100.0 100.0 Source: UNICEF Philippines 2022. If the number of observations is large enough, the same analysis can be disaggregated specifically to the most common type of households (for example, a couple with two children or a person of pension age in a single-person household). The analysis can be extended to measure the nonfood share if a certain minimal food expenditure is reached, thus enabling comparison of total costs with and without a person with disabilities. The Philippines study, for couples with two children, shows than when households have the same food consumption, those with a child with disabilities have a 22 percent higher total expenditure than those without a child with disabilities. The feasibility of this type of analysis depends on the availability of a national survey that collects adequate data: detailed consumption expenditure data, proxies for disability and official disability status (when applicable), questions on disability-specific expenditure items, and information for calculating potential subsidies. Its main advantages are that it uses existing data, which keeps the costs relatively low; it identifies the most common sources of disability-related expenses; and if the survey includes enough observations, it can estimate overall disability-related expenditure compared with households without members with disabilities. The method can also be combined with the standard of living (SOL) method, because both rely on the same data. Section 3, “Standard of Living Method,” includes a discussion on the resources, costs, and duration of the CPA method combined with the SOL method. Methodological guidelines on assessing household disability-related 22 costs and their implication for participation Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 2.2 Limitations and Recommendations for CPA CPA in its basic form does not estimate the additional disability-related expenditure but identifies differences in spending priorities and needs of persons with disabilities and their families, highlighting the types of disability-related expenses they incur. Disability-related needs will be underestimated if a country has few services for persons with disabilities, if services are available only in the capital or major cities, or when disability-specific items are unavailable or difficult to access. In such cases, the analysis should be supplemented with a qualitative component to better understand the situation of persons with disabilities. Analyzing survey data can offer insights into unmet needs. Many surveys include some information on unmet needs in health and education, using questions about school enrollment or unaddressed health problems. Descriptive comparisons of unmet needs between people with and without disabilities can reveal important differences. Box 2.1 presents examples of these questions and how they can support the analysis. Another relevant indicator sometimes available in these surveys is the employment or labor force participation rate, which can be disaggregated by disability status. The other substantial limitation of CPA is the difficulty in distinguishing needs by type of disability. Most often, analyses differentiate only by disability severity and thus estimates typically reflect average costs across disability types. This limitation could be addressed if the survey contains enough observations by type of disability and the sampling is representative, but this is rarely feasible. Surveys often lack a good sampling frame for persons with disabilities, and even when such frames exist, they are rarely considered in these surveys. Section 3, “Standard of Living Method,” explains that when conducting this type of analysis, involving the country’s national statistical office for input can help improve questionnaire design in future surveys. Box 2.1:   Measuring Unmet Needs Using Common Questions in Household Surveys Household surveys used to construct consumption expenditure aggregates and monitor living standards often include separate sections on access to education and health. These sections usually collect person-level information and related expenditure when people attend school or address their health needs. Education Each household member is asked about their highest level of education. For people in the typical age range for education (for example, between ages 3 and 24), the survey asks whether they are currently enrolled in school. If not, it asks when they stopped attending school and why. These data allow comparisons of enrollment rates between people with an official disability status or functional difficulties and those without, across different age groups. This analysis reveals gaps in school enrollment and provides a measure of unmet need for basic education. If observations of persons with disabilities are limited, and the survey includes when the disability started, the analysis can also consider gaps in educational attainment among persons with disabilities above the age of basic education. For example, based on the Household Socioeconomic Survey for Mongolia in 2022, analysis of school attendance by disability shows that 37.4 percent of children ages 5–17 with functional difficulties were not enrolled in school, compared with only 3.6 percent of other children. In Georgia, using its 2022 Household Income and Expenditure Survey, 26.2 percent of children with an official disability status were enrolled, compared with 2.5 percent of other children. Failure to enroll in or complete basic education indicates an unmet need. Health To identify health expenditures, surveys often ask whether household members experienced a health problem or needed a health consultation. If they answer “yes,” a follow-up question asks whether they sought care and, if not, why. These responses, along with the reasons for not seeking care, can be used to assess unmet health needs and compare between people with and without disabilities. Methodological guidelines on assessing household disability-related 23 costs and their implication for participation Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix The Kyrgyz Republic’s 2014 Integrated Household Survey asked whether people needed medical assistance in the past year and if so, whether they did not use health services when they needed them, including the main reason why. The survey also asked whether they had been referred for hospital or inpatient treatment but did not use those services, and why. This information can be used to compare the level of unmet needs between people with and without disabilities. Among people aged 15 and older who received a disability pension, 20 percent reported at least one time that they did not use needed health services, but among other people in the same age group, the figure was 14 percent. For hospital or inpatient treatment, the rates were respectively 19 percent for those receiving a disability pension and 4 percent for others. Source: Original analysis of microdata conducted for this report. In all three cases, household surveys are nationally representative and have been collected by the national statistical offices. Methodological guidelines on assessing household disability-related 24 costs and their implication for participation 03. Standard of Living Method KEY MESSAGES The SOL method estimates disability-related costs by comparing the consumption expenditure of households with and without a member with disabilities, assuming they reach the same standard of living, which is measured using a nonmonetary indicator. The method assumes that higher consumption expenditure leads to higher household living standards, and that households with members with disabilities need to incur higher expenses to reach the same standard of living of other households. However, disability-related expenses—additional costs required to overcome impairments and environmental barriers—are not improving people’s welfare compared with other households but instead are “regrettable necessities” incurred because of the circumstances faced by persons with disabilities. Berthoud, Lakey, and McKay (1993) developed and implemented this method of estimating disability-related costs in the United Kingdom, but it was after an analysis by Zaidi and Burchardt (2005) that the method started to be used in other countries. Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Simply comparing the overall consumption expenditure of households with and without a member with disability could show that on average, households without a member with disability have a lower level of expenditure just because they are poorer than other households. However, an adequate comparison between households must consider the living standard achieved, which the SOL method does and thus overcomes the income constraint and sometimes of related unmet needs. However, if disability- specific items and services are unavailable (even at relatively higher levels of incomes), the method will underestimate the full cost of what is really needed. Therefore, these costs are lower-bound estimates linked to the type of goods and services available in the country. Figure 3.1 shows the stylized relationship between living standards and consumption expenditure and the differences related to disability. In households with a member with disabilities at each level of living standard, the disability-related cost is the difference between the expenditure for households with disability and other households. Figure 3.1: Standard of Living and Disability-Related Costs Without disability Measure of standard of living With disability Disability-related cost Income or Consumtpion expenditure Source: Original figure for this publication; World Bank and UNICEF 2025. The calculation method is inferential and does not identify the specific items that are disability-related expenditures,9 especially if the calculation is based on income rather than consumption expenditure. However, when consumption expenditure data are available, the SOL assessment can be combined with CPA to provide additional information on the sources of the disability-related expenditure. Disability-related expenditure can vary with the standard of living, so no single measure applies, and it could change depending on the standard of living achieved. The representation is only schematic and does not assume a specific pattern in how disability-related expenses change across different living standards. These expenses could increase, decrease, or remain constant. Still, expecting actual disability-related expenses to be minimal at low living standards is plausible, because persons with disabilities and their families cannot afford them, and thus they fail to reach minimum living standards. As households reach adequate living standards and begin to participate in society, expenses clearly increase. At higher living standards, it becomes harder to predict whether expenses continue to increase. This can be tested empirically and might depend on the degree of functional difficulties. 9. The estimate is even more inferential when the living standards measure accounts for unmet needs, so a lower living standard might not be associated with actual disability-related expenditures but instead the failure to incur them. Methodological guidelines on assessing household disability-related costs and their implication for participation 26 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix If the standard of living measure captures the household’s full economic situation and accounts for unmet needs, the estimate of disability-related expenses can go beyond actual spending to reflect the broader costs of disability. Even when total expenditure is low and no specific disability-related expenses are recorded, unmet needs could lower the household’s living standard compared with others with the same expenditure level. Although the living standards measure often includes household assets and housing conditions, it could include other variables such as the subjective assessments of economic situation, unmet needs, food security indicators, and others. A more comprehensive living standards measure allows the method to move beyond actual expenditures to estimate the broader costs of disability. A recent study in the Philippines provides a simple but powerful example of SOL analysis. The study’s analysis considered the case of households of couples with two children (the most common type of households with children in the Philippines). The living standards indicator captured the number of generic unmet minimum needs,10 and as figure 3.2 shows, the average consumption expenditure increased as the number of unmet minimum needs decreased from three or more to none (figure 3.2). However, at each number of unmet needs, the average consumption expenditure of households with at least one child with a disability card is higher than households without a member with a disability card. Figure 3.2: Consumption Expenditure by Living Standard and Disability Level for Two-Parent, Two-Child Households in the Philippines Unmet minimum needs None One Two Three or 4000 14000 2000 8000 12000 10000 16000 6000 Other households Household with disability card Consumption expenditure (units) Source: UNICEF Philippines 2022. Note: PHP = Philippine peso. 10. The unmet needs included in the analysis are the following: at least one child of school age is not being enrolled in school, at least one household member is declaring to have forgone health visits/treatments, lack of household members with health insurance, household is below the acceptable standard for food consumption, the household reports hunger episodes, below minimum standards for access to water and sanitation facilities, and without adequate rights to live in the dwelling where they reside. Methodological guidelines on assessing household disability-related costs and their implication for participation 27 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Estimating disability-related expenditure using the SOL method requires analyzing a regression model, where the general equation can be written as follows: SOL=β0+ β1 ln(consumption) + β2disability + βiHHtypei + βjcontrol variablej + ε (3.1) where SOL is the measure of standard of living; consumption is the variable estimating the household consumption expenditure, which is generally included in the formula expressing it in logarithmic terms; disability represents the variable or variables used to capture a proxy on how disability is present in the household; HHtype provides dummies of the main types of household, based on size and composition; and several other control variables. The disability related cost is estimated using the coefficient of the consumption variable and that of the disability measure. The next sections explain each element in detail. Methodological guidelines on assessing household disability-related costs and their implication for participation 28 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 3.1 Choice of Standard of Living Indicator The nonmonetary measure of living standards is critical in assessing disability-related expenditure. It should respond to the measures of consumption expenditure or income and capture households’ welfare but ideally reflect the well-being of the person with disabilities. This methodology assumes that households spend part of their resources on meeting disability-specific needs, which take precedence over other needs, implying that households with members with disabilities reach lower living standards than others at the same level of expenditure. If the living standards measure includes unmet needs, households also show lower living standards when disability-related expenditure does not occur. Therefore, living standards indicators vary in their sensitivity to this equalization process. A household wealth indicator based on asset ownership and housing characteristics is the most common household welfare measure in SOL models because of the availability of such data in household surveys, not from a specific methodological requirement. Household asset accumulation occurs over time and could be affected by the onset of a disability. However, in larger households, the impact of disability is spread across more members,11 making it less likely to affect asset ownership, which tends to be shared. Other indicators of household welfare include deprivation, social participation, and subjective assessments of a households’ ability to meet economic needs or to save. Because these indicators are measured at the household level, they risk overlooking the needs of the person with disabilities. As previously noted, intrahousehold discrimination and social norms about activities and integration of persons with disabilities can also influence outcomes. To address this, including person-level indicators such as unmet medical needs or enrollment is useful if they correlate strongly with household consumption expenditure and income. Combining different indicators into a single index requires a defensible weighting of the different components. Most commonly an index is data driven and computed using principal component analysis. Some researchers have applied multidimensional poverty measures in this analysis (using normative weights assigned to different life dimensions), but these require some careful evaluation, especially when they show weak correlations with consumption expenditure (see Pinilla-Roncancio [2023] for a discussion on using and interpreting non-monetary measures in this type of analysis). Still, different indicators can be combined using principal component analysis to test whether the resulting indicator correlates with consumption expenditure. In LMIC, asset indexes (using ownership of durable items and housing conditions) are most common, but more comprehensive indexes can be constructed by including subjective indicators of economic condition, food security, and unmet needs in the principal component analysis. 11. In some contexts, a larger household size could be an endogenous response to the presence of a member with a disability. Methodological guidelines on assessing household disability-related costs and their implication for participation 29 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 3.2 Consumption or Income The questionnaire design often determines whether consumption or income is used, based on which measure better suits the country context—in low informality contexts, income is used most often, and consumption is used where informality is high. The choice is usually based on practical rather than theoretical considerations (Deaton and Zaidi 2002; Mancini and Vecchi 2022). In most countries of Africa, eastern Europe, and Asia, typically use consumption expenditure (which is available in household survey data) as the welfare indicator, but income is the prevalent measure in Latin America and OECD countries. In the context of analyzing disability-related costs, thechoice between using income or consumption as a measure requires careful consideration. Because income fluctuates more than consumption, its relationship with a nonmonetary indicator is likely less robust. Consumption smooths needs over time through savings, dissavings, and borrowing. For example, if a shock lowers household income, consumption might be better protected through debt. How income and consumption relate to living standard depends on whether the living standards measure reflects short-term or long-term effects (such as the ability to save versus accumulated assets), the life cycle of the household, and the stage of disability or how long ago it occurred.12 In some cases, the welfare indicator excludes components of consumption expenditure, including health- related spending. When analyzing disability-related costs, the consumption aggregate should include health expenditure. This requires attention to cases that might appear to be outliers but instead reflect genuine expenses incurred by persons with disabilities. Therefore, ensuring that health and other disability-related expenses are included in the consumption aggregate depends on how consumption expenditure is calculated. When consumption information is collected, SOL analysis can be accompanied by an analysis of consumption patterns. 12. Some studies have tried to address endogeneity between income and consumption measures and disability by using panel data (see, for example, Cullinan, Gannon, and Lyons [2011]). Panel data control for the endogeneity, correct for possible unobserved effects, and allow calculating the effect of longer-term disability. However, these studies are data-demanding and have been implemented almost exclusively in high income countries. Methodological guidelines on assessing household disability-related costs and their implication for participation 30 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 3.3 Functional Form of Consumption Expenditure The functional form used to transform consumption expenditure or income is independent from the choice of income or consumption. The most common approach is to transform the aggregate using a logarithmic form, which is usually the one that best fits the data and the relationship between the nonmonetary measure of living standards and the consumption and income aggregate. However, the choice of functional form influences interpretation of the interaction between the coefficient of the disability variable and the aggregate. Using consumption in logarithm implies that the incurred disability-related cost is measured as a ratio. If households have the same attributes and reach the same living standard, then the following is true: β1 ln(consd ) + β2 = β1 ln(consnd ) (3.2) where the suffix d stands for households with a member with disability and nd for other households. Therefore, the ratio of the consumption of households with members with disabilities compared with other households is: (3.3) The relationship between living standard and the logarithm of consumption is linear, which means that the disability-related costs are estimated to be a fixed proportion of household consumption. In absolute terms, these expenses rise as the living standard increases from low to high (figure 3.3). Figure 3.3: Logarithmic Functional Form and Size of Disability-Related Costs Measure of standard of living Ln (Consumption) Consumption Source: Original figure for this publication; World Bank and UNICEF 2025. However, the evolution on the size of disability costs is imposed by the relationship between consumption and living standard. To better understand how disability costs vary as living standards increase, either introduce in the regression interaction terms between disability and consumption terciles or quintiles,13 or estimate independently the regression for households with and without members with disabilities. Both approaches require significant samples of households with persons with disabilities and ensuring that no bias is present, because the characteristics of persons with disabilities are different across the 13. Estimating separate regressions for different levels of consumption is considered inappropriate. They are truncated regressions known to produce biased estimates. Methodological guidelines on assessing household disability-related costs and their implication for participation 31 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix consumption distribution.14 When data observations allow, analyzing the evolution of costs between households at different stages of income levels should be done considering homogeneous household types. Disability-related costs as a ratio can be used in estimating equivalence scales for persons with disabilities and correct poverty measures to account for disability costs. Box 7 discusses the different methods used to correct poverty measures for the presence of disability-related costs. Box 3.1:   Correcting Poverty Measures for Incurred Disability-Related Costs Different methods can try to correct poverty measures for disability-related costs. One approach estimates the absolute value of disability-related costs (at the mean or median) and subtracts them from the overall consumption expenditure aggregate or income aggregate before calculating the welfare indicator used for poverty measurement. For example, Asuman et al. (2021) and Palmer, Williams, and McPake (2019) use this method, which is the same as increasing the poverty line by the percentage of disability-related costs when estimating poverty among households with a member with disability. Surveys also use this approach when trying to directly estimate the disability-related expenses and compare the disposable income of households with and without disability only after subtracting such expenses (the goods and services used method). However, this method risks applying the average costs of disability across households of different sizes and thus overestimates the disability cost for large households and underestimates it for small ones. This approach is similar to the analysis of impoverishing health expenditure, in which case the actual health expenditure is considered, ignoring possible unmet needs resulting from people not seeking health treatment. Another approach is to try to estimate the specific equivalence scale of the person with disability, which requires differentiating the analysis for households of different size and composition, but this approach is theoretically superior in describing the poverty profile of households with persons with disabilities. For example, Zaidi and Burchardt (2005) and Carraro et al. (2023) follow this method. Disability Variable The survey can capture proxies for disability in different ways. Depending on the type of information available, disability can be represented as a probability score or as a dummy variable, but even if using a dummy variable, distinguishing cases of moderate and severe disability is useful (box 3.2).15 14. For example, large households are likely to have high expenditures, and small households have low consumption. A finding that disability- related expenditure is higher at lower consumption levels may reflect differences in household composition and not a real decline in expenditure from poorer to better-off households. 15. The concept of disability is not well defined and clear cut and thus contains measurement errors. Some studies have tried to define disability considering a latent variable, which underlines the manifestation of observed effects, such as functional difficulties. The same can be said for the living standards measure. Morciano, Hancock, and Pudney (2015) have used structural equation modeling with latent variables specification to reduce the effect of measurement errors and theoretically produce more robust estimates of coefficients to calculate disability-related costs. Methodological guidelines on assessing household disability-related costs and their implication for participation 32 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Box 3.2:   Possible Disability Proxies and How to Measure Disability Severity Using the Washington Group set of questions to ask about functional difficulties is becoming best practice, but not all researchers do so and sometimes must use alternative proxies, depending on the information the questionnaire provides. The following are examples of possible candidate questions that are alternative disability proxies: ö Questions capturing whether people suffer from chronic or long-term illnesses. ö Information about people who are limited in their daily activities (for at least six months) because of a health issue. ö Questions about functional difficulties but on a smaller set of domains than those the Washington Group recommends. ö Some surveys might ask whether persons in the household have a disability or not or if they have certain impairments/ disability (e.g. blindness or deafness) or if there is anybody in the household who requires additional care due to illness, disability or old age. Disability Severity The Washington Group questions usually infer disability severity from the person’s response and the degree of functional difficulty experienced across the different domains (some difficulty, a lot of difficulty, or cannot do at all). Within this set of questions, “a lot of difficulty” or “cannot do at all” in the self-care question can be interpreted as a sign of very significant disability, because it identifies people who are likely to need high levels of human support. Severity is then progressively lower among those who answer “cannot do at all” in other functions, those who answer “a lot” in at least one function, and those who answer “some” in two functions. Other questions can also include some degree of severity, or severity can be identified based on the combination of difficulty reported in different functions. If different types of questions are asked, they could be combined into a single index using principal component analysis. Source: World Bank and UNICEF 2025. Household Types and Control Variables In the regression model, controlling for other factors that can affect the living standard is necessary, but ideally it should be independent of the consumption expenditure variable. One essential variable is household size or dummies capturing the main types of households.16 If the number of observations allows, running separate regressions by household types is preferred, because detecting an impact of disability-related costs in small households would be easier than in large ones. Including household size or type as dependent variables corrects for the role of different household types on the living standards measure but does not address the relationship between household consumption and disability, which will be calculated for the average household type. However, with just one household member with disabilities, the disability-related expenditure will be felt more in small households when the cost is expressed as a percentage of household income or expenditure. Other control variables commonly included in the analysis are geographical (urban and rural areas or different regions of the country), and the household head’s age (to capture the different stages of the household life cycle) and education. Endogeneity could affect the basic regression model, and many of the control variables could be correlated with disability and with consumption and income. As with all household survey data, measurement error is inevitable. These factors could affect the correct estimation of the key coefficients. Measurement error in the consumption variable is likely to underestimate the consumption coefficient, which could overestimate the measure of disability costs. Exceptionally high estimates of disability-related expenditure without support in different consumption patterns should be interpreted with caution. Good practice is to check for multicollinearity and test the results of the model with and without control variables that are highly correlated, for example, without the education of the household head. 16. Using dummies for household types allows for possible nonlinear effects between the living standards measure and the household size. Methodological guidelines on assessing household disability-related costs and their implication for participation 33 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 3.4 Implementing the SOL Method Implementing the SOL method requires a quantitative survey collecting detailed information on consumption expenditure or income of the household, proxies for identifying persons with disabilities in the household, and indicators for nonmonetary measurement of the household living standard— information that surveys by national statistical offices usually collect to measure monetary poverty. When the surveys collect consumption expenditure, the same surveys can be used for CPA. Using existing datasets, the SOL method can be implemented in three main phases: ö Preparation phase: involves verifying that the country has a household survey with the information needed to conduct the study, recruiting consultants and forming a working group, and mapping current programs aimed at reducing disability-related costs. ö Data analysis phase: depends on the type and level of information collected in the household questionnaire, which can include imputing the subsidies received by persons with disabilities, assessing programs’ effectiveness (in-kind and cash benefits), the CPA and assessment of unmet needs, calculating SOL measures, the SOL regression, and calculating the effect of disability-related expenditure on monetary poverty measures. ö Reporting phase: preparing the report and disseminating the findings and recommendations. The depth of SOL analysis depends on having relevant information from the questionnaire and on the number of observations of households with a member with disabilities (a function of sample size and disability prevalence in the country). The number of observations determines whether the SOL regression can assess disability-related expenditure considering the severity and type of disability across different consumption expenditure levels (terciles or quintiles) and for different household types. In some cases, the possibility of increasing the number of observations and combining data from additional household surveys from different years could be explored. The survey questions might allow measuring unmet needs in access to education and health, and some surveys also contain information on time use, which would allow estimates of the time that household members spent caring for persons with a disability, thus examining an important cost and use of household resources that has significant gender differences (box 3.3). Box 3.3:   Time Use Modules and Disability The 2021–22 Household Budget Survey in Bosnia and Herzegovina includes a module about the use of time of all household members ages 15 and above. People reported the frequency with which they care for their children and grandchildren, engage in cooking or household chores, care for elderly or other family members, engage in voluntary activities, and engage in sports or leisure activities. Analysis carried out showed that for households of two adults, daily care was provided in 40% of cases when the other member has significant functional difficulties, reaching 80% when the person has difficulty in self-care compared with less than 5% when there is was nobody with functional difficulty (Carraro, 2024). Similarly, the Household Income and Expenditure Survey in Kiribati has a time-use module, in which each household member between ages 12 and 65 reports the number of hours spent in the preceding week engaged in chores (cooking, washing clothes, gardening, and so on), childcare, caring for other dependent household members, producing handicrafts, attending special family events, and in religious activities. Time-use modules can be used to compare the time use across households of similar size and composition but with or without a household member with a disability. It enables identifying who within the household provides care to the person with disability and the effort required and also in relation to disability severity. The analysis can also be linked to employment among working-age people in the household and the indirect costs of disability. Source: World Bank and UNICEF 2025. Methodological guidelines on assessing household disability-related costs and their implication for participation 34 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Implementing the analysis often involves hiring a consultant with expertise in this type of assessment and necessary statistical knowledge. The overall cost of the study, using existing datasets, would be approximately $15,000, which includes a quick mapping of existing programs, survey data analysis, and preparing and presenting the report. If focus group discussions and active participation of organizations of persons with disabilities are included, as recommended, the cost would increase to about $20,000. The study would take at least one month if carried out continuously, but a more realistic timeline is up to three months. A separate document17 delves into the details of implementing this method, providing practical guidance on the different stages and offering template terms of reference and programming code segments for the data analysis. 17. Access the operational guidance on implementing the SOL approach at https://www.worldbank.org/en/topic/disability Methodological guidelines on assessing household disability-related costs and their implication for participation 35 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 3.5 Limitations and Recommendations for the SOL Method The SOL method has clear, strong theoretical advantages because it can potentially overcome problems of lower incomes among persons with disabilities by basing the estimate on reaching a certain living standard. However, the estimate of needs through disability-related expenditure is underestimated if disability-specific items and services are unavailable in the country, and if current social norms discriminate against persons with disabilities and their inclusion in daily life (they are hidden from society and have low levels of participation). Ensuring that the living standards measure responds to the unmet needs of persons with disabilities can partially overcome this limitation, but in such cases, the monetary cost estimate remains a statement of disadvantage rather than an accurate representation of the disability cost. The SOL method has been used widely in recent years, but the analysis rarely makes full use of available data to assess disability-related expenditure in a comprehensive way. Most studies rely only on estimates from regression models. Household surveys conducted by national statistical offices serve multiple purposes and are a valuable public good and efficient use of resources. In most countries, analysis of the surveys should include the following: ö An assessment of the level of inclusion of persons with disabilities and whether disability-specific benefits and services exist. If inclusion is low and programs are absent, disability-related expenditure will most likely be small, making the analysis insufficient to understand needs. ö An assessment of generic unmet needs compared with others and indicators allowing assessment of indirect disability-related costs, such as time use. Some surveys include questions in their education or health section that can be used to analyze unmet needs, but if not, the study could be an opportunity to advocate for their inclusion, because they are relevant to the entire population, not just persons with disabilities. Adding time use sections helps assess the time household members spend caring for the person with disabilities but also a broader assessment of unpaid care work distribution. Households rarely incur costs for paid human assistance, but the cost is instead reflected in the household members’ time. ö A combination of the CPA and SOL methods, which makes it possible to estimate overall disability- related spending and identify what drives those costs. Even within some income constraints and lack of information or availability of services, consumption patterns can provide insight into what is available and reflect current social norms. A study based on existing household surveys could also include qualitative interviews to illustrate quantitative findings and provide context. Even when disability-specific goods and services are available, the main limitation of CPA-SOL analysis is that it usually does not provide separate estimates by disability type, and results represent overall average estimates, or simply distinguish expenditures for two or three groups based on the degree of functional difficulties. Methodological guidelines on assessing household disability-related costs and their implication for participation 36 04. Goods and Services Used KEY MESSAGES The goods and services used (GSU) method identifies and measures disability-related costs by analyzing consumption expenditure. Questions focus on expenses which are known to arise because of disability and are considered additional to what other households normally spend. By focusing on persons with disabilities the method can explore in greater detail the types of goods and services they use and the related expenses. It also considers different needs based on impairment types, thus showing the diverse mix of goods and services used by disability type. This method is used commonly for determining disability-related expenditures because of its relative simplicity (and its long history). However, it has significant weaknesses. It does not correct for income constraints and identifies only additional expenditure.18 It asks persons with disabilities to compare their own expenses with those of hypothetical households without a person with disabilities, requiring them to make assumptions about what others spend. For this reason, the expenditure measured with this type of survey is also known as a subjective cost assessment. An alternative is to conduct a full assessment of consumption expenditure and compare it with similar consumption measures available for the general population. However, this approach also has significant challenges with comparisons because of possible differences in questionnaire design and the survey timing and implementation. A key factor in these assessments is the representativeness of the sample and whether the analysis accounts for households’ income and living standards relative to other households. Failure to do so contributes to variability in the estimates of disability-related expenditure, which can either underestimate or overestimate them and limit meaningful comparison with other households. Box 4.1 summarizes the challenges and disadvantages of implementing this type of survey based on experiences in the United Kingdom. 18. In some areas of expenditure, persons with disabilities could have lower needs than others. To estimate overall disability-related costs, the lower—or “negative”—costs must be subtracted from the additional costs. Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Box 4.1:   Surveys of Persons with Disabilities in the 1970s and 1980s in the UK Several small-scale surveys were conducted in the UK in the 1970s focused on people with a specific impairment. For example, in 1975, Hyman interviewed 56 wheelchair users about additional costs related to transportation, diet, heating, clothing, laundry, phone, and medical supplies. The largest survey of persons with disabilities was conducted in 1985 by the Office of Population Censuses and Surveys (OPCS), interviewing 11,158 adults with disabilities. The survey examined the prevalence of disability; the respondents’ financial circumstances; and services, transportation, and employment. It calculated an average disability-related expenditure, disaggregated by disability severity (10 groups) and impairment type. The survey also showed that disability-related expenses were higher for people with higher incomes than for other households, and that 24 percent of adults documented having unmet needs in relation to disability expenses. Because the estimates of disability expenses derived by the OPCS surveys were lower than disability organizations expected and lower than the amount provided by disability benefits, the Disablement Income Group conducted another small-scale phone survey in 1988 of 87 persons with disabilities using a more narrowly focused questionnaire exclusively on disability- related costs. This survey’s results showed significantly higher disability-related costs estimates than the OPCS survey. The analysis of disability-related expenses and a comparison between households with and without members with disabilities was possible from the national Family Expenditure Survey and a follow-up survey in 1986–87. The comparison of expenditure between households with and without members with disabilities provided an estimate similar to that of the OPCS survey, but when applying the Standard of Living methodology, the resulting estimate was in between the OPCS and the Disablement Income Group estimates. Source: Summary based on the discussion in Berthoud, Lakey, and McKay 1993. The GSU methodology’s strengths are in measuring differences in expenses across disability types. For example, a recent UNICEF-supported study in Armenia interviewed 1,300 caregivers of children with disabilities selected from the national registry of children with disabilities. The study sampled from the five official types of disabilities (mental disorders, hearing impairment, visual impairment, locomotor disability, and chronic diseases) through implicit stratification. The survey used video calls interviews to collect information about household income and the costs of caring for the child with disability. The costs were divided into 15 categories: special food, special clothing, assistive devices, medication, hygiene items, health care services, medical rehabilitation services, social rehabilitative services, transportation costs, education costs, other costs for education, home adaptations, personal assistant, and other costs (for example, entertainment, toys, and summer camps). The analysis calculated the expenditure incurred for the child with disability as a percentage of total household income and analyzed it across the five different categories of disability. Despite the method’s strength, comparing expenses for a child with disabilities with the expenses incurred for other children and then comparing them with national estimates is extremely difficult. However, costs can be compared with the value of the disability allowance, and the study provides valuable information on some aspects of the implementation and effectiveness of the disability allowance. Other countries use disability registries or social registries with information on functional difficulties as a sampling frame to survey persons with disabilities. The surveys collect data on incurred expenses and barriers to accessing assistive products and other essential services to assess the implementation and effectiveness of disability-specific programs. For example, in the Philippines, people with a recognized disability can receive a disability card that entitles them to discounts for some expenses: health care and medicines, transportation, and a stated amount for groceries. The Philippine Statistics Authority uses the Family Income and Expenditure Survey to monitor poverty and living standards, but this survey lacks information on who is registered as disabled and can use a disability card. Instead, a nationally representative survey that sampled specifically children with a disability card collected such information, which was used to assess the effectiveness of disability-specific subsidies and cash transfers (box 4.2). Methodological guidelines on assessing household disability-related costs and their implication for participation 38 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Box 4.2:   Assessing the Effectiveness of Current Interventions for Persons with Disabilities National surveys can provide simple measures of the effectiveness of current interventions for persons with disabilities and their families, as long as the survey includes enough respondents with disabilities and collects data on cash benefits and specific services provided to them. Example of indicators include coverage (such as the percentage of persons with disabilities receiving a cash benefit), the average level of support received across programs, and whether the support is a progressive or regressive transfer. When possible, coverage estimates should consider both people with an official disability status and those likely to have a disability based on reported functional difficulties. However, because the national survey in the Philippines does not collect data on specific subsidies linked to a disability card, a special survey was needed that sampled specifically people with a disability card and assessed the percentage of children with disabilities who had officially recognized status. The survey found that a large percentage of children with official status did not have a disability card, but among those who did, the value of the subsidy—through the discounts they were entitled to receive—benefited better-off groups disproportionally. The finding raised concerns about the fairness of the policy and its failure to adequately support children with disabilities in rural areas and low-income households. Figure B4.2.1 shows the distribution of the subsidy from the concessions system to people with a disability card by quintile, highlighting that the top two quintiles receive most of the subsidy, and the bottom quintile receives very little. Figure B4.2.1. Distribution of Disability Card Subsidy by Consumption Quintiles, among Children with Disability Card (% of total subsidy, total=100) Poorest quintile 5.9 Second 14.6 Third 10.6 Fourth 25.8 Richest quintile 43.1 Source: UNICEF Philippines 2022. Methodological guidelines on assessing household disability-related costs and their implication for participation 39 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 4.1 Implementing the GSU Method The GSU has been implemented using quantitative surveys in most cases, but it can also be conducted using qualitative interviews, especially if the goal is to demonstrate the needs of people with different types of impairments. The first step is to define the study objective in the country context and determine how it could complement available information. Conducting a GSU study to complement a CPA-SOL analysis or a goods and services required analysis is recommended, because promoting this study as stand-alone research is more difficult. For example, if the objective is to understand the different needs across disability types, a series of focus group discussions—each with people representing different impairments—could accomplish this goal. However, if an assessment of policies for persons with disabilities does not exist, a quantitative survey with a representative sample could fill that gap. A disability registry and contact phone numbers, if available, would make phone surveys an inexpensive and effective option when the questionnaire has a specific focus. The method can be applied selectively to certain types of disabilities. If using a quantitative survey, the approach should include probability sampling and a rigorous sampling strategy. Box 4.3 outlines possible sampling approaches for surveys of persons with disabilities. The survey design should also consider existing national statistical office survey questionnaires to ensure comparability and identify improvements that could be adopted in national surveys. For example, including and testing questions on unmet needs, time use, and access to state programs (such as cash benefits and services) is strongly recommended. When using qualitative methods, the selection of case studies should reflect what is known about the types of disabilities in the country and consider available definitions and statistics on disability severity. The questions should cover the same topics covered in a quantitative survey: sources of disability-related costs, goods and services used and their expenses, unmet needs, time use, and access to state-provided cash transfers and services. As with the other methods, forming a working group is recommended with representatives from relevant line ministries, the national statistics office, and organizations of persons with disabilities. The group will provide information to consultants and be the main platform for discussing the study’s findings. The study design should be grounded in a thorough understanding of the existing government programs for persons with disabilities, and if other study components have not already done so, these programs should be mapped as part of the study. Costs for this type of study can vary significantly depending on its scope and the sampling method best suited to the country context. For quantitative samples (excluding exercises involving nonrandom samples), a GSU survey can be conducted on a relatively low budget if a phone survey is possible, which requires a national disability registry. In such cases, the study could cost about $25,000 and take from four to six months to complete. However, sampling frames that allow phone surveys are uncommon, especially in LMIC. The budget would increase sharply if the study requires in-person surveys with population samples and screening questions to identify people with functional difficulties. Methodological guidelines on assessing household disability-related costs and their implication for participation 40 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Box 4.3:   Probability Sampling and Persons with Disabilities Several small-scale surveys were conducted in the UK in the 1970s focused on people with a specific impairment. For Probability samples prevent selection bias in estimates and allow for calculating sampling errors. They require that everyone in the target population has a non-zero probability of selection. These samples usually rely on a sampling frame (a list of the target population), but sampling methods such as clustering or multistage sampling can overcome the need for a full sampling frame (Henry 1990). For persons with disabilities, common sampling frames include: ö A disability registry, if available. ö A social registry with disability information. ö A recent census, if it includes questions on disability and suitable proxies and accessible data (though this is often difficult). Disability and social registries should be assessed to determine if they are comprehensive, updated regularly, and have adequate population coverage. When no sampling frame is available, clustering or multistage sampling is the typical alternative. Using population data from general population surveys or size-related population subgroups allows researchers to select geographical areas or clusters (primary sampling units), then list and screen households in only the selected clusters to identify people and households with the required characteristics, who are then randomly sampled—an approach that works well, because the number of persons with disabilities often correlates strongly with the population. Additionally, geographic stratification can reflect regional differences in disability prevalence. When registries are incomplete, combining sampling approaches is possible. Clusters can be selected using probability proportional to population size, and then within these clusters, both households identified through the registry and other randomly selected households can be included. For households not in registries, the Washington Group questions can identify households with at least one person with functional difficulties as a proxy for disability. Such studies do not need to be nationally representative. A regional focus can still ensure rigor in the selection and representativeness of the sample. Source: World Bank and UNICEF 2025. Methodological guidelines on assessing household disability-related costs and their implication for participation 41 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 4.2 Limitations and Recommendations for the GSU Method The GSU method’s estimates of disability-related costs are likely to be biased, because they fail to correct for unmet needs and lower incomes of persons with disabilities. The approach’s main strength is the ability to disaggregate estimates by type and severity of functional limitations. A sample of enough persons with disabilities allows assessing the effectiveness of current programs targeted to them. A survey using the GSU method is recommended primarily for complementing data from other methods and could be necessary if a national household survey does not capture information on disability. Phone surveys can obtain valuable information at a lower cost and are possible in countries with disability registries and available contact information. A quantitative analysis using this method could also be used to develop and test questions that could be included in national statistical office household surveys. Qualitative assessments using this method can complement an analysis based on the CPA-SOL method. Methodological guidelines on assessing household disability-related costs and their implication for participation 42 05. Goods and Services Required KEY MESSAGES The goods and services required (GSR) method aims to go beyond revealed preferences, which rely on observing consumption choices and usage, and instead measure what persons with disabilities actually need. A key element of this approach is defining the purpose of the required elements, with the objective of ensuring that persons with disabilities participate equally in the society where they live and achieve a minimum or decent living standard. The GSR method can be implemented in several ways: ö Quantitative GSR: asks persons with disabilities what they consume, what unmet needs they have, and what they would need to spend to meet those needs. ö Comparative qualitative GSR: an extension of the approach used for the general population under expert-driven or consensual budget standards, removing the assumption that all household members are in good health and considering household types with at least one member with disabilities. ö Exploratory qualitative GSR: a simplified version of the comparative GSR but with no need for a comparison group, focusing on disability-specific needs or other substantial costs that are clearly disability-related. Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 5.1 Quantitative GSR The quantitative GSR approach identifies the most common disability-related goods and services during the design phase. For each item, respondents are then asked: (i) how much they use and spend on it, (ii) whether it met their needs, and (iii) if not, how much more they would need and what that would cost. This approach aims to account for the income constraint that limits consumption of disability-related items and relies on declared preferences to estimate the cost of these goods and services. This approach has several limitations. The assessment of need is subjective and not necessarily linked to a specific standard of living or participation target. Even among disability-related items, not all assistive devices or services are considered necessary. Subjective needs often reflect people’s expectations, which can vary by socioeconomic status—for example, people who are relatively better off frequently report higher needs. Although this issue cannot be fully resolved, survey questions should clarify what is meant by “need” and explore whether additional items are useful, necessary, or essential to performing a certain function. Declared preferences might differ significantly from actual behavior if the income constraint were removed. People’s responses to hypothetical questions often reflect different incentives or thought processes, which is why the study should also collect information on household income. The most significant limitation is the assumption that respondents know about the goods and services available to meet their specific needs. For example, if people are unaware of a particular assistive device or cannot identify it in a list, they would have difficulty determining whether they need it or estimating its cost. Therefore, the approach will likely work better where information is widely available and relevant goods and services are usually accessible and known to people, which is why the approach has mainly been used in relatively developed economies. For example, Morris et al. (2024) applied it in the United States, collecting data on both out-of-pocket expenses for disability-related items and unmet needs. The study also gathered income data and found that higher income was associated with fewer unmet needs. Implementing this approach requires a quantitative survey, which could be a population survey that oversamples persons with disabilities and their families or a survey focused exclusively on them (ideally including various disability types). With the latter, the analysis faces the same constraints as the GSU approach: the absence of a comparison group from the rest of the population (see section 4.2). Methodological guidelines on assessing household disability-related costs and their implication for participation 44 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 5.2 Comparative Qualitative GSR The comparative qualitative GSR approach for persons with disabilities and their families depends on an existing assessment of needs for the general population. Budget standards define the basket of goods and services required to participate in society and estimate its cost. Budget standards are calculated for specific household types (such as working-age single person, pension-age single person, working-age or pension-age couples, and couples with one or two children) and needs are assessed for all life dimensions. However, they typically assume that people are in good health. To assess the disability-related costs, this assumption is removed, and at least one household member is considered to have a specific disability. Then, different scenarios are developed based on different types of disability and assessing the associated needs. Disability-related costs are estimated and compared against the benchmark of the same household structure without a member with a disability. This process identifies the kinds of costs involved and highlights which items are removed and others are added based on household needs. As with general needs assessment, this approach can be led either by experts or through a consensual budget standard approach. Both ways have been used widely in many countries, including some developing ones19, but extending this approach to disability has been implemented in only a few advanced economies (Belgium, Ireland, and the United Kingdom and to some extent in New Zealand)20. Figure 5.1 outlines the research process used to develop consensual budget standards. It starts with baseline household types without any members with disabilities and proceeds in three steps: 1. Orientation groups define minimum living standards needed to participate in society, and social scientists then validate the consistency of the key assumptions. 2. Item selection involves creating lists of goods and services needed to meet the minimum living standards, achieved by taking an imaginary tour through each room of the house and linking rooms to necessary items (kitchen to food, living room to leisure, and so on). Social scientists use the lists to calculate costs and ensure that they meet standards, such as nutritional adequacy. 3. Focus group discussions (FGDs) review and revise the proposed lists and budget. Contested items are discussed until participants reach consensus. Each FGD includes members of households that match the target household type (for example, couples with a primary school–age child). Participants might also complete expense diaries and inventories of possessions to help them focus on their needs, but the goal is to define the needs of a hypothetical household, not the actual households in the FGDs. The baseline household type is then modified to include a member with a specific disability (for example, a child with an intellectual disability in households made by a couple with a child). The three stages are repeated to adjust the baseline budget to fit the needs of the new household type until a consensus is reached. This approach provides a detailed assessment of how the goods and services required change—both additions and subtractions—based on the presence of a disability. As figure 5.1 shows, the research process is time consuming and expensive. 19. See Deeming (2020b) for a collection of research on minimum income standards and reference budgets. 20. For early applications of this approach in the UK, see Middleton, Ashworth, and Walker (1994) and Dobson and Middleton (1998) on children and children with severe disabilities, respectively, and Smith et al. (2004) on adults with disabilities. Ongoing work at the Centre for Research in Social Policy at Loughborough University has estimated and updated minimum income standards regularly since 2008 (Davis, Hirsch, and Padley 2018), and Hirsh and Hill (2016) have estimated disability-related costs and assessed them against a benchmark. In Ireland, the Vincentian Partnership for Social Justice produced significant research on budget standards and minimum essential standards of living. As in the UK, MacMahon and Thornton (2020) have calculated them for various household types, and MacMahon, Boylan, and Thornton (2022) calculated disability-related costs against those benchmarks. Belgium has long developed reference budgets for different household types (Penne et al. 2020), but the work has recently focused on applying this approach for children with disabilities (Gijbels et al. 2023). In New Zealand, similar research on disability-related costs was conducted at the Disability Resource Centre and University of Auckland, but these estimates were compared loosely with budget standards developed in Australia (Wilkinson-Meyer et al. 2010). Methodological guidelines on assessing household disability-related costs and their implication for participation 45 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Figure 5.1: Use of Group Discussions to Determine Consensual Budget Standards Baseline household types Member with disabilities Step 1: orientation groups agree on Successive FGDs generate consensus on additional costs of disability Successive deliberative FGDs involve different people in each step language, concept of minimum Step 1: standards, and case studies identify additional costs using the agreed budget as a baseline Experts check assumptions Step 2: task groups negotiate lists of minimum requirements Step 2: Participants complete review results in step 1 and diaries and inventories produce possible additions or changes Experts calculate and cost the budget Step 3: check-back groups consider lists Step 3: and agree to final budget Review results in stage 2 and produce final consensus Experts make final amendments Source: Adapted from Middleton 2000, Bradshaw et al. 2008, and Hirsch and Hill 2016. Methodological guidelines on assessing household disability-related costs and their implication for participation 46 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 5.3 Exploratory Qualitative GSR Exploratory qualitative GSR (or exploratory GSR) simplifies the comparative qualitative GSR and is better suited for LMIC. Its policy objective is to identify possible areas of intervention. This approach uses qualitative methods, but its exclusive focus on persons with disabilities means that no comparison group is needed, and it considers only disability-specific goods and services and other substantial, clearly disability- related costs. Thus, the approach does not consider situations in which they and their families have lower costs than other households. The objective is not to produce accurate cost estimates but to understand the range of these costs and their structure: assistive technology, personal assistance, rehabilitation, health care, education, transportation, and other costs. For simplicity the costs identified through this approach are considered disability-specific, even though other substantial costs are included. The other objective is to investigate the heterogeneity of needs (goods and services) and their costs across different disability types. The term “exploratory” indicates that it is not a complete assessment of disability-related costs. Because the approach cannot provide nationally representative estimates, it focuses on illustrative cases that reveal needs across different policy areas. The focus on disability-specific items is intentional. In countries where such goods and services are limited or unavailable, the costs are often the first area in need of intervention to improve the lives of persons with disabilities. Like the comparative qualitative GSR, the exploratory approach combines expert opinions with FGDs involving persons with disabilities or parents of children with disabilities to identify required goods and services. The approach is participatory, involving stakeholders throughout the study to build ownership and understanding of the disability-related costs. Because exploratory GSR is better suited for LMIC, and earlier versions of this approach were implemented in Georgia, (UNICEF 2023); Peru (UNICEF 2025); Tamil Nadu (CIP 2024), and South Africa (NDSD 2015), the rest of this section focuses on this method. A companion paper provides operational guidance and tools for its implementation.21 Four key groups implement exploratory GSR studies, each with specific roles: ö Advisory committee: Composed of government officials, organizations of persons with disabilities, other nongovernmental organizations, and donors, the committee guides the study’s scope and scale. ö Researchers: The group leads the study, facilitates communication between the committee and experts, and conducts FGDs. ö Experts panel: The panel includes professionals and persons with disabilities (or parents of children with disabilities) who support the identification of required goods and services, helping to address constraints in assessing disability-related costs such as lack of information, limited access, low incomes, and intra-household distribution issues. ö Persons with disabilities (or parents of children with disabilities): This group participates in FGDs to validate the lists of goods and services. Persons with disabilities and parents of children with disabilities can contribute to the study as experts and in FGDs because of their first-hand experience of living with a disability. 21. Access the operational guidance on implementing the exploratory GSR approach at https://www.worldbank.org/en/topic/disability Methodological guidelines on assessing household disability-related costs and their implication for participation 47 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Four Phases of the Exploratory GSR Approach Exploratory GSR studies are implemented in four phases: study preparation, conducting FGDs, reconciliation and costs estimation, and reporting. The study uses a strong participatory approach involving government officials and organizations of persons with disabilities from the beginning and throughout all phases, which means active participation in understanding the study framework, obtaining the initial results, and explaining findings and their policy implications. The participatory approach promotes understanding and ownership of the results, which helps build political support and strengthens stakeholders’ ability to advocate for more inclusive policies. Managing expectations requires explaining the study’s purpose and boundaries: the aim is limited to understanding the structure and range of disability-related costs and identifying possible areas of intervention. Figure 5.2 outlines the four phases of the exploratory GSR approach. The sections after it explain the phases and describe key methodological aspects of the process and the expected results of the study.22 Figure 5.2: Exploratory GSR Implementation Overview Phase 1: preparation Advisory Experts panel Develop key study parameters: groups of persons with disabilities (age, committee professionals, disability types, low and high support needs), definition of participation government service providers, Map the availability of programs, benefits, and services (public and representatives, and persons private), propose categories of goods and services analysis OPDs, NGOS, with disabilities Propose preliminary list of goods and services, justification and frequency of use Prepare the FGD guidelines Phase 2: Discussions FGDs Validate the list of goods and services required and propose For each type of disability and low modifications and high support need, balanced for Validate what existing programs provide gender, age, socioeconomic background, urban and rural Provide examples of use of items and rationale Phase 3: Costs estimation Reconciliation and calculation of Categorize goods and services, reviewing and agreeing inclusion and costs' structure intensity and frequency Experts and researchers Identify unit costs per item (market research) Construct illustrative case studies (by type of Phase 4: Reporting Develop report and present findings Outputs: lists of goods and services required by disability type and Researchers need, extent of coverage in existing government programs, costs for illustrative cases Presentations and report Source: Original figure for this publication; World Bank and UNICEF 2025. Note: FGD = focus group discussion; GSR = goods and services required; NGO = nongovernmental organization; OPD = organization of persons with disabilities. 22. Additional guidance on the implementation of this method including on the different stages and offering template terms of reference, questionnaires and other useful material is available at https://www.worldbank.org/en/topic/disability Methodological guidelines on assessing household disability-related costs and their implication for participation 48 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix First Phase: Study Preparation and preliminary lists of goods and services The first phase is study preparation, which requires input from the advisory committee, researchers, and the experts panel to define the study’s scope, map existing disability services and benefits, create a preliminary list of goods and services, and develop FGD guidelines. Experts and Their Role. Experts in this study include both professionals with extensive experience working with persons with disabilities (such as clinical professionals, occupational therapists, psychologists, and pediatricians) and persons with disabilities or parents of children with disabilities who are well-connected within disability communities (for example, through organizations of persons with disabilities or associations of parents of children with disabilities). Ideally, the expert panel should include a balanced number of professionals and knowledgeable persons with disabilities. For each type of disability, both groups should be represented, because they bring different perspectives and contributions to the assessment. Defining distinct groups of persons with disabilities. To explore the diversity of needs, the study must define the different groups of persons with disabilities (impairment, level of support needs, children and/ or adult, rural and/or urban.) it will include and establish clear criteria for what constitutes low or high support needs. Definitions should align with national classifications and be compatible with available data on prevalence and identification. A lack of prevalence data for disability types makes it difficult to interpret cost estimates at macro and national levels, underscoring the importance of integrating qualitative and quantitative methods. Where prevalence data are unavailable, linking to official disability classifications is still useful because of the possibility that administrative data will become available in future following these definitions. Defining the Reference Standard of Living and participation. The study must define what “participate in society” means in the national context. This requires identifying the living standard and participation that people are expected to achieve, considering common societal expectations across life dimensions— education, frequency of travel, health care, and social participation—and selecting appropriate terminology in the local language. In some countries, terms such as “minimum” level of participation or income are associated with subsistence or survival; alternatives such as “dignified life” or “basic living standard” might better reflect the intended meaning (Padley and Davis 2020). For persons with disabilities, the emphasis must be on equal participation in society, but this must be assessed relative to the living standards and participation that others in that country typically achieve. (See box 5.1 for examples of how living standards have been defined in consensual budget standards studies). A clear, context-specific definition of the target level of participation is necessary to identify the goods and services needed. Mapping Current Programs. The assessment should map all existing disability support programs and services, including in-kind transfers and targeted subsidies (direct cost reduction), cash transfers that provide income (indirect support), and in countries with limited or no government services or programs, community-based rehabilitation services. Identifying where rehabilitation exists—whether government supported or widespread—can inform service development. In countries with limited disability services, estimating costs can be challenging. Prices may be unreliable or reflect market immaturity, requiring cautious interpretation. Additionally, costs would not reflect the situation if provision of goods and services is scaled up with the possibility of reducing prices. Similarly, this framework cannot account for possible price changes in volatile environments and where technological improvements can affect costs significantly. Preparation of preliminary lists. This is the primary responsibility of the panel of experts which usually split in groups, each focusing on a specific disability type. In each group the list of goods and services is Methodological guidelines on assessing household disability-related costs and their implication for participation 49 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix developed starting by considering people with low support needs and then those with high support needs. While each group develops the list independently, it is useful to have regular meetings where all experts share their progress, and possible challenges. These meetings should be facilitated by the researchers. This ensures that the same standards and approach are applied across disability types. The experts together with the researchers should also develop the guides for the focus group discussions. Box 5.1:   Example Definitions of Participation and Income Required to Achieve It United Kingdom: A minimum standard of living includes food, clothes, and shelter, but also having enough resources to access opportunities and make choices that enable participation in society. Mexico: A dignified life includes meeting basic needs—food, housing, and clothing—along with access to work opportunities, education, health care, and leisure. It also involves living in a stable, secure environment that encourages social connection and inclusion. South Africa: A decent living level allows people to participate fully in society in ways they choose. It includes personal belongings, social and family networks, a belief system, housing, local services, and social opportunities. It reflects a basic standard that everyone should attain, not a life of luxury. Source: Padley and Davis 2020. Second Phase: Conducting FGDs The second phase involves conducting FGDs with distinct groups, organized by disability types and low and high support needs. The FGDs validate the preliminary list prepared by experts, identify the types of goods and services commonly available, and provide examples explaining the need for different items. FGDs review the list of items that experts developed by proposing additions or deletions based on relevance, not cost. The FGDs’ main value lies in sharing useful examples and personal experiences, especially about access to items and government services. FGDs also help clarify what adequate personal or human assistance means and how families struggle with this need, especially when formal services are limited or unavailable. Personal assistance is difficult to price, but it remains essential. Third Phase: Reconciliation and Costs Estimation The third phase is a reconciliation exercise that resolves any disagreements between the experts’ input and the FGDs findings, and the items are organized consistently across policy categories and disability types. Items are then costed through a specific analysis of prevailing prices. This phase produces illustrative cases for each disability type, with selected items and their frequency documented for each case. Price Research. Researchers conduct a survey to determine the prices of each item in the list, collecting price ranges, type of provision (private or government), and source of information (website, store, or official government document). Constructing Illustrative Cases. Illustrative cases are stylized representations sometimes based on real examples from the study but designed to reflect common situations within each disability type and level of support needs. The cases are needed to analyze disability-related costs, because the list of goods and services and their prices for different disability types do not reflect actual disability-related costs. Individuals within each group need only some of the listed items, often at varying frequencies. This is why their costs must be standardized to a monthly equivalent, which involves adjusting each item’s price based on how often it is needed, then scaling to a uniform monthly value. Methodological guidelines on assessing household disability-related costs and their implication for participation 50 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Fourth Phase: Reporting The final phase produces reports, presents the findings, and delivers the outputs: a mapping of current efforts to reduce disability-related costs, the list of goods and services required for each disability group (type and low and high support needs), the cost of each item, and the illustrative cases and their costs. Study Outputs. The main outputs of the exploratory GSR are: ö The map of existing support for persons with disabilities. ö The list of goods and services required for participation for each disability type and low and high support needs groups (including costs, availability, state support for each item, and the item’s purpose). ö Illustrative cases showing the differences in costs across disability groups (types and low and high support needs) and the cost structure. Box 5.2 presents fictional examples of outputs from the analysis of illustrative cases. Box 5.2:   Fictional Examples of the Type of Outputs from Illustrative Cases In some countries where this type of study has been conducted, outputs from the analysis of illustrative cases have helped understand the diversity of disability-related costs and supported the identification of areas for policy development. Figure B5.2.1. Disability-Related Costs in Illustrative Cases across Disability Types and Low or High Support Needs Type 1 (name a) Type 2 (name b) Type 3 (name c) Type 4 (name d) Type 5 (name e) 450 180 360 270 90 State Household Sources: Original figure for this publication; World Bank and UNICEF 2025. Note: Being a fictional example, the five types represent the chosen disability types in the study, and names represent the case studies within that disability type. Figure B5.2.1 clearly shows the different range of disability-related costs that exists between low and high support needs and across disability types. Understanding that responses cannot be uniform and that needs differ greatly can be useful. Methodological guidelines on assessing household disability-related costs and their implication for participation 51 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Figure B5.2.2. State Support and Household-Borne Disability-Related Costs for High-Need Support Cases Type 1 (name a) Type 2 (name b) Type 3 (name c) Type 4 (name d) Type 5 (name e) 450 180 360 270 90 High Low Sources: Original figure for this publication; World Bank and UNICEF 2025. Note: Being a fictional example, the five types represent the chosen disability types in the study, and names represent the case studies within that disability type. Figure B5.2.2 clearly shows the gap between what the state provides and the needs of different types of disability. Although needs differ, the state support is relatively uniform. Figure B5.2.3. Structure of Costs for High Support Needs in Illustrative CasesSources: Original figure for this publication; World Bank and UNICEF 2025. Type 1 (name a) Type 2 (name b) Type 3 (name c) Type 4 (name d) Type 5 (name e) 100% 40% 80% 60% 20% Assistive devices Human assistance Rehabilitation Others Sources: Original figure for this publication; World Bank and UNICEF 2025. Note: Being a fictional example, the five types represent the chosen disability types in the study, and names represent the case studies within that disability type. Understanding the structure of costs is necessary to go beyond the overall figure and see for what support is required in the different types of disabilities. Methodological guidelines on assessing household disability-related costs and their implication for participation 52 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 5.4 Limitations and Recommendations for the GSR Method The GSR is the only method that explicitly estimates the full cost of disability, including unmet needs. It does so through direct questions, estimation of the cost of unmet needs, expert input, and consensus on what goods and services are necessary. Exploratory GSR is best suited for LMIC and relies on experts from professional fields and representatives of persons with disabilities. Its main limitation is the inability to produce nationally representative estimates. It offers only illustrative cases and assumes uniform preferences. The method also struggles to reflect regional differences in services availability and quality and in determining their costs. Implementing it in countries with limited disability services is especially challenging. National experts might have a narrow view of the list of goods and services, leading to cost underestimates. In such cases, international experts can offer valuable guidance and inputs. Pricing in underdeveloped markets also increases the risk of inaccuracies. Despite these limitations, the GSR method provides information on costs composition and structure and their diversity across disability types, which is policy-relevant for determining interventions. This makes it a good tool to identify the scope of potential programs that could support disability inclusion. Where possible, GSR should be complemented by a quantitative CPA-SOL analysis. Methodological guidelines on assessing household disability-related costs and their implication for participation 53 06. Measures to Address Disability-Related Costs: Developing a Support Package KEY MESSAGES Previous discussion explored the different methods for assessing disability-related costs, but the attention now shifts to the scope of the policy implications. Disability-related costs affect the participation of persons with disabilities. A key question is: what types of interventions can reduce these costs? Investments in infrastructure and the environment can reduce barriers and the need for additional costs, but the focus here is specifically on social protection programs. Findings from disability-related cost assessments offer a foundation for designing support packages, but translating them into effective policy requires additional steps. The process should start with a review of existing programs to improve their ability to promote equity. Support must align with individual needs, ensuring access for all persons with disabilities. Programs may include cash transfers—either to offset disability-related costs or provide income support—and in-kind transfers such as subsidized health care, assistive products, and social services. Studies help identify appropriate support levels and types, showing that no single measure covers all needs. A well- designed package must combine instruments, balance equity and agency, and reflect each country’s capacity (Cote, Knox-Vydmanov, and Lippi, 2024). Cash transfers are common instruments, as are in-kind benefits such as health care and provision of assistive products and services. However, understanding the size and structure of disability-related costs is necessary to inform policy intervention design and support package development. No single intervention can address all the needs of persons with disabilities. Instead, a comprehensive package of different interventions should be tailored to each country’s circumstances. Insight into the advantages and disadvantages of each policy instrument can help in understanding trade-offs and the rationale for creating an efficient and effective combination of programs. Cash transfers usually compensate for average disability-related incurred expenditures, but in-kind support can better address diversity of needs, and government interventions can provide some disability- specific goods and services. This section reviews the role and design of cash transfers and then examines the role of in-kind provision. It highlights the complementarity of different interventions and the need to develop a comprehensive package addressing the diversity and level of needs of people with disability. It then discusses how to develop a response package under very different scenarios that mimic the conditions in many countries. The section provides an overview of aspects to consider rather than detailed instructions on how to develop the interventions package. Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Figure 6.1: The main categories of social protection instruments, types of schemes and related functions across the lifecycle EXAMPLES OF TYPES SCHEMES ACROSS THE LIFECYCLE CATEGORIES OF MAIN FUNCTION INSTRUMENTS CHILDHOOD WORKING AGE OLD AGE Poverty assistance cash transfer; cash for work Unemployment; maternity; Child grant/family sickness; parental leave benefits INCOME benefits SECURITY Old age pensions Disability related income CASH Caregiver benefits replacement benefits BENEFITS Disability costs allowance / top up schemes Child disability benefits Third person support / caregiver benefits COVERAGE OF DISABILITY Concessions RELATED Health insurance/cost coverage including early intervention; COSTS AND rehabilitation; and assistive devices; free or discounted public HEALTH CARE transport; and utilities; tax exemptions; food subsidies COSTS IN-KIND BENEFITS Services Health care; nutrition; case management community care and support; rehabilitation, assistive technology, and personal assistance schemes; interpreters; respite care; counselling; point to point transport; housing assistive products maintenance Early childhood Economic inclusion/empowerment CONNECTIONS development; programmes; decent work programmes; return INTERLINKAGES TO OTHER childcare; to work programmes; women’s empowerment; SERVICES education protection services; financial inclusion services Source: Adpated from the joint guidance in Cote, Knox-Vydmanov, and Lippi, 2024. Methodological guidelines on assessing household disability-related costs and their implication for participation 55 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 6.1 Cash Transfers to Address Disability-Related Costs Many countries use cash transfers to support persons with disabilities. For example, ILO (2024) states that worldwide, 176 countries provide statutory disability cash benefits, but coverage varies considerably— nearly 100 percent in Europe and North America but only 8 percent in Africa, for example.23 However, little is known about whether these benefits cover disability-related costs adequately. Cash benefits for persons with disabilities can have two distinct objectives: income support and compensation for disability-related costs. Cash transfer programs can address one or both of these objectives. Cash transfers for caregivers also address indirect costs by considering the caregivers’ income needs. The four main types of cash benefits are: ö Disability insurance pension (providing income maintenance). This benefit, often called disability pension (or retirement pension), is based on social insurance contributions. The benefit’s value depends on the specific social insurance program, but eligibility requires formal employment or participation through voluntary contributions. The objective is to maintain income, and it is not linked to disability-related costs. When people are of working age, this benefit is a disability pension, but when they reach retirement or pension age, the benefit converts to a retirement pension. ö Disability benefits addressing disability-related costs. Examples are disability allowances for persons with disabilities regardless of their contributions, increased child benefits for children with disabilities, or additional payments for those who need care and support such as third person support or personal assistance allowances. ö Minimum income programs (poverty-targeted benefits). These support low-income households to reduce the extent of extreme poverty in the country. Eligibility is based on income or proxies. Where disability is considered in addition to household minimum needs, support should reflect the additional disability-related costs. When adjusted this way, these benefits address both poverty and disability- related costs. ö Caregiver benefits. These benefits support caregivers directly —often family members and most commonly parents of children with disabilities—who cannot work because of caregiving duties. In some cases, caregiving years count toward pension eligibility. These benefits recognize the cost of human assistance and compensate for lost income. In many countries, eligibility for disability benefits (pensions or allowances) is still solely associated with inability to work, which goes against the spirit of the United Nations Convention on the Rights of Persons with Disabilities. Although some persons with disabilities cannot work due to very high support needs, many would be in better position to do so if they received assistance to meet disability costs related to seeking and keeping work in the informal or formal economy. Considering the impact of disability-related costs, policies should seek to facilitate inclusion and participation by removing the barriers persons with disabilities face and proving them adequate support. Even where benefits are designed to address disability-related costs, their value often does not reflect those costs. Poverty-targeted benefits also rarely adjust for the household’s specific needs if it includes persons with disabilities. It is important to acknowledge that cash transfers are essential instruments in 23. The International Labour Organization calculates coverage only of working-age persons with disabilities and considers whether they receive a disability-specific cash transfer, such as a disability pension or disability-related early retirement. Methodological guidelines on assessing household disability-related costs and their implication for participation 56 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix supporting persons with disabilities, but they cannot alone address the diversity of disability-related costs which requires a combination of in cash and in kind support. Estimating disability-related costs—especially by support needs—helps governments assess the adequacy of disability benefits and improve the design of minimum income programs benefits. However, a disability allowance can cover only part of disability-related costs but not all of them. Estimates based on spending for an equal living standard or those based on goods and services used could guide the level of support that a cash transfer can provide. An exploratory GSR allows for disaggregating disability-related costs to identify which can be covered by a cash allowance. When using the different methods in the same country, findings and estimates can be triangulated to determine common elements. If results are consistent and robust, a range of estimates can be obtained using the SOL, GSU, and GSR methods to compare with the current benefit levels to determine their adequacy. In countries without disability benefits, estimates can be used to propose them. However, these studies will never provide a unique estimate but instead will produce evidence of these costs and a cost range to guide policy decisions on setting the value of cash allowances, which also involves political considerations. Actual expenditure levels are especially relevant in determining the values of cash assistance relative to available disability-specific goods and services. A disability allowance can help stimulate demand, potentially increasing the use of available services. Study findings can also inform the design of cash allowances relative to the degree of functional difficulties and the number of persons with disabilities in the household. Many findings support varying the benefit based on functional limitations or the need for substantial human assistance. Evidence also shows that costs increase with the number of persons with disabilities in the household, which supports person-level instead of household level allowances. Therefore, benefit amounts should vary by degree of care and support needs, such as mobility limitations or need for substantial human assistance, thus triggering a higher allowance. Setting these amounts is part of a political process and country context–specific. For example, some countries might start with lower amounts and increase them gradually to the desired level, depending on budget estimates and the number of persons with disabilities entitled to the allowances. Still, where possible, structural features—such as differentiation by support need—should be set from the beginning and aligned with the country’s disability assessment system. Cash allowances can address only part of the disability-related costs, but they should focus on the additional costs incurred. Ideally, benefits should be available to all persons with disabilities, regardless of their age. For example, a disability pension as income maintenance can be paired with a payment addressing disability-related costs. In countries with a disability certification criteria that distinguishes different level of official disability status (e.g mild, moderate, or severe), estimates of disability -related costs could guide cash benefit levels: no cash benefit for those with mild disability status, a basic amount for those with moderate disability status , and a higher amount for with severe disability status. The value could also vary by age (children or working adults or older persons). The cash transfer would compensate for some of the disability-related costs but not for insufficient income, which is addressed through social insurance programs that cover employed people who become disabled; through support to those without social insurance (for example, they work in the informal sector or become disabled before engaging in work); or when the household lacks the resources to support the household’s minimum needs. Methodological guidelines on assessing household disability-related costs and their implication for participation 57 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Some countries do not provide support to all registered persons with disabilities but only to those with a relatively low income. When comparing with other people or households in such cases, the definition of “relatively low income” must account for the disability-related costs. When cash transfers are targeted to households based on poverty or economic need, eligibility assessment must account for the additional needs of persons with disabilities, which affects which households qualify and the amount of the cash transfer. How to do so depends on the type of needs assessment and whether the country has an allowance for disability-related costs. If the needs assessment requires a direct means test and the country lacks an allowance to cover disability related costs, household income should be divided by an equivalized household size, considering the extra needs of any household member with a disability. Another way is to subtract disability-related costs from the household income before conducting the assessment. If the benefit is then calculated based on the gap between the household income and an eligibility threshold, the transfer will reflect disability-related needs. Alternatively, if a universal disability allowance already covers those costs,24 the allowance should be excluded from the income test, and then ignore the disability-related costs. If the assessment is conducted through some proxy means test, the underlying model must include the additional needs of persons with disabilities either in the welfare indicator or the household needs component, and then proxies must include the presence of households with members with disabilities. When proxy means tests are used to target cash transfers, the benefit amount is usually fixed and calculated at the household level or relative to the number of household members. Within the benefit amount’s present structure, the amount can be adjusted to reflect the extra needs of households with a member with disabilities. Table 6.1 summarizes the eligibility assessment. For this exercise, the SOL method might be more appropriate than others, because estimating eligibility rules and their modeling often uses household survey data. Therefore, SOL estimates of a specific disability equivalence scale will be consistent with the same data and based on comparison with other households, which ensures the consistency of approaches across all household types. Table 6.1: Eligibility Rules to Poverty Targeted Benefits Accounting for Disability-Related Expenditure Type of eligibility Presence of disability allowance assessment No disability allowance With disability allowance Income test (household income minus disability-related 1. Disability allowance is equal to disability- expenditure) is lower than eligibility threshold related costs. Therefore, (household income minus disability allowance) is lower than eligibility threshold 2. Disability allowance is different from disability-related costs. Therefore, (household income including disability allowance minus disability-related costs) is lower than eligibility threshold Proxy means test Household needs assessment must reflect The model estimating the means of disability-related expenditure (model estimating the household must detract from the household income or consumption must dependent variable the disability allowance, subtract disability-related expenditure from and the proxies must include the presence dependent variable, and proxies must consider of disability in the household. the presence of disability in the household). Source: Original table for this publication; World Bank and UNICEF 2025. 24. This assumption requires specific testing. Even if the allowance is initially adequate, it might not retain its real value over time or continue to cover disability-related costs. If the allowance does not increase regularly for inflation, its adequacy should be reviewed regularly. Methodological guidelines on assessing household disability-related costs and their implication for participation 58 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Box 6.1 provides examples from countries that have adjusted their eligibility assessment to consider the effect of disability-related costs. In poverty-targeted benefits, the income eligibility threshold can affect the value of disability-related costs. A threshold set at relatively low levels risks underestimating the indirect disability-related costs. Box 6.1:   Adjusting Proxy Means Tests to Account for the Needs of Persons with Disabilities Several countries—such as Georgia, Moldova, Mongolia—have adjusted the proxy means test model to account for the additional needs of persons with disabilities and to identify households eligible to receive cash transfers. All three countries have estimated specific equivalence scales for persons with disabilities and in some cases differentiate between children and adults with disabilities and disability severity. All three used the standard of living method to estimate the equivalence scales. In Moldova, where the income test determines both eligibility and eventual amount of the benefit, the social support benefit reflects the additional needs of persons with disabilities. In Georgia, a recent proposed revision of its Targeted Social Assistance program includes an adjusted amount of support if a person in the household has a recognized disability status. Source: World Bank and UNICEF 2025. Table 6.2 summarizes the types of cash benefits for persons with disabilities and their different rationales. Disability cash allowances address only part of the overall disability-related costs, covering only disability- related expenditure. Even when cash allowances consider the degree of support needs, they cannot address the diversity of needs across disability types or a lack of disability-specific services, which is why complementary policies should be developed through in-kind provision of specialized goods and services. Table 6.2: Cash Benefits and Their Role in Addressing Disability-Related Costs Cash benefit Main objective Link to disability-related costs Disability pension from social insurance Income maintenance Amount independent of disability- program related costs Disability pension from social Compensation for disability-related Amount should cover disability- assistance program costs, lack of income, or both related costs to compensate for their disadvantage Poverty-targeted benefit Ensure that everyone has a minimum Eligibility for and amount of benefits income and reduce extreme forms of must consider disability-related costs, poverty to ensure equity Caregivers’ benefit Support people for their loss of Address the human support income and recognize their valuable component of disability-related costs contribution and requires identifying persons who need substantial human assistance Source: Original table for this publication; World Bank and UNICEF 2025. Methodological guidelines on assessing household disability-related costs and their implication for participation 59 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 6.2 In-Kind Benefits In-kind benefits include direct provision of services (health and health-related services, personal assistance, social services support, employment support, and so on), concessions for persons with disabilities, and subsidies for certain disability-specific goods and services. These benefits reduce a household’s costs to acquire goods or services, but the scope of the government intervention may be limited to lowering the cost of an item for a specific group of people or could extend to ensuring the availability of the item. In many cases, the government’s intervention addresses market failures: consumption of a certain item produces positive outcomes that markets fail to recognize, or the government’s intervention makes the market viable. Providing cash or reducing the price of disability-specific items might appear to achieve the same objective or lowering disability-related costs, but in-kind benefits often support the creation of services that otherwise would not exist. This especially matters for goods and services with low demand, and government intervention ensures their availability. In-kind benefits allow tailoring support to individual needs, enabling better responses to different requirements based on the type of disability support needs and degree of functional difficulty. The government may provide these services directly, manage them, or contract with private providers. Concessions can also apply to generic goods and services like public transport, in recognition that persons with disabilities generally face higher expenditures and lower incomes compared to their non-disabled peers. When designing in-kind support, evidence from GSU and GSR studies helps identify the types of goods and services needed by different groups of persons with disabilities and levels of support needs. These goods and services often involve high costs that cash benefits cannot cover and require professional need assessments on a case-by-case basis. Health-related services are a main area of in-kind support. The type of support depends on how the country finances health expenditure, but in all cases, careful assessment of needs and available funding is required to ensure that services are provided adequately. In systems that rely partly on health insurance, governments often cover premiums for persons with disabilities, but the health care package must meet their specific needs. Even when some health services are publicly funded or insured, often copayments are required, and some programs provide reduced fees for persons with disabilities. Few countries include assistive technology in a standard health package. In some cases, separate systems managed by social protection ministries and agencies manage this provision, and in others, the availability of assistive technology is inconsistent, left to voluntary organizations to provide, or is completely unavailable. Government involvement can reduce the costs of assistive technology significantly.25 Depending on the intervention’s stage of development, the aim should be to create a priority list of assistive products that can be expanded gradually. The findings from exploratory GSR on discrepancies between needed and currently provided products can guide this process. Additionally, the delivery method (reduced subsidized price, minimal contribution, or reimbursement programs) requires careful assessment to identify and address access barriers. Another type of in-kind support includes subsidies or concessions for expenses that persons with disabilities incur because of a disadvantage (for example, besides health-related copayments, these may apply to transportation, utilities, hygiene products, and diapers). A key concern in implementing such concessions is the lack of information about their effectiveness, requiring careful assessment of the 25. Government intervention is especially needed in developing speech-to-text and text-to-speech software in local languages for assistive products that often are available only in English or other widely spoken languages. Methodological guidelines on assessing household disability-related costs and their implication for participation 60 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix concession’s design and how it is administered. This occurs with improper budgeting and monitoring and poor implementation. For example, cases in which the service provider should absorb the cost or the business that provided the discount should be reimbursed through reduced taxation pose considerable risks to program implementation. If the institution bears the costs, it can discourage service provision. Similarly, if businesses are reimbursed through reduced taxation, only formal enterprises can participate, limiting access to those able to use the service. Partial discount instead of full subsidy may also favor better off people who have the capacity to pay the non-subsidized portion. Concessions on public transport, though very useful, may favor people in main urban areas where such formal transport services are available. Box 6.2 outlines the key elements for designing a concessions system. Box 6.2:   Designing the Structure of Concessions Two different policy objectives need to guide the design of a concession: the consumption level achieved (meaning how well the policy helps persons with disabilities to reach the optimal consumption of the item), and the distributional impact (how the subsidy is received across the income distribution). These two policy objectives involve a trade-off, and the design needs to consider the characteristics of the item: whether people have different needs for the item or whether similar needs can be reasonably assumed. For example, medical needs can vary depending on health condition and type of severity of impairment. Transportation needs also differ by functional difficulties with some people able to use regular public transport, some needing accessible transport and others requiring point to point transportation. By contrast, needs for basic food and utilities tend to be more uniform. When assessing needs, the focus must be on the disability-specific needs of the person, not the person’s or household’s ability to pay, which relates directly to income. Subsidy Design When Needs Differ Supposed person A needs 10 items, person B needs five, person C needs three, and persons D and E each need one. A total of 20 items are needed. Assume that a subsidy can cover only 15 items. Several subsidy options are possible: proportional support (discounted item price), equal support (equal quantity provided free of charge), or equal loss (equal financial burden). Under proportional support, the item’s price could be reduced to 25 percent of the actual cost. The benefit each person receives reflects their consumption. However, if someone cannot afford even the subsidized cost, they might not reach their optimal consumption. People who are better off are more likely to reach optimal consumption therefore receive a larger share of the subsidy. Under equal support, the first three items are provided free of charge, and any additional items must be paid in full. This design penalizes those with the highest needs, who would end up paying the highest amount, and it could also encourage unnecessary consumption among those with lower need. Under equal loss, the financial burden is shared equally among all consumers, who all pay the full price for the first item, and then all other items are provided free of charge. Although each approach could be structured to have the same total cost, their distributional impacts differ depending on the welfare of the recipient and in situations when people cannot afford the portion of the cost not covered by the subsidy (see González [2020] for a discussion of health copayments in Europe). In some cases, proportional support might yield better results than equal support, and in other cases, equal loss could perform better than proportional support. Determining the most effective model requires empirical analysis that considers the subsidy budget, the type of goods subsidized, and the income distribution of potential beneficiaries. Subsidy Design When Needs Are Similar When the item benefits everyone equally, final consumption depends on the level of subsidy and people’s ability to pay. In these cases, the subsidy’s distributional impact depends on whether lower-income or higher-income individuals are more likely to consume the item. However, if needs are the same, the objective becomes promoting consumption of the item with positive distributional impacts. The most effective design in such cases is a capped subsidy that ensures equal support to all users. Source: Carraro et al. 2024. Specific services that could support persons with disabilities and their families include respite services, personal assistance, community care and support, counseling, and support services related to access to early childhood intervention and education (including early education) and work participation. The availability and development of these services depend on the country’s social care services and network of social workers and employment offices. These services may be provided directly or contracted from Methodological guidelines on assessing household disability-related costs and their implication for participation 61 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix nongovernmental organizations or the private sector. The design of these services should help strengthen the participation of persons with disabilities in the labor market and in education, creating important connections with other sectors. Preferential Tax Treatments Many countries offer preferential tax treatments for persons with disabilities. Some programs provide direct support, such as income tax reductions, which compensate for disability-related costs associated with employment. Another common approach is to reduce the costs of some disability-specific items by reducing or exempting taxes on the item (such as value added, sales, or import taxes). Tax benefits for firms that hire persons with disabilities are designed only as incentives to employ persons with disabilities. However, favorable tax treatment for companies that invest in inclusive work environment can also help reduce disability-related costs. When a country adopts preferential tax treatments, it should avoid generic programs that might be seen as charitable and instead link the benefits clearly to the rationale of offsetting disability-related costs and other related interventions (Cremin 2018). Some tax treatments also recognize impact if unpaid caregiving provided by family members, often women, by allowing years spent caring for a person with disabilities to count toward pension entitlements, which is equivalent to the government paying social insurance contributions during those years. Methodological guidelines on assessing household disability-related costs and their implication for participation 62 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 6.3 Designing the Package: Pros and Cons of Different Instruments Ideally, a holistic assessment of an individual’s needs would guide creation of a tailored personal budget. The person could then use this budget to meet their disability-related costs, within defined limits. Some OECD countries use this model, though it is complex to administer. A similar outcome can be achieved by providing different packages of diverse measures: cash transfers to cover common expenditures and allow flexibility for an individual’s specific needs, concessions to reduce the cost of disability-specific items or offset overall disability costs, and services that address the high costs disability-specific items faced by some persons with disabilities. The support package would adjust to levels of need so that all persons with disabilities would have access to affordable health care, assistive products, rehabilitation, inclusive services, and economic empowerment programs based on demand, not on official disability registration. However, registration should be required to access concessions. Registration that includes documented disability-related costs would trigger disability allowances, which would be paid regardless of employment status or receipt of other cash benefits. People with very high support needs could receive higher or additional cash support and specific services, such as personal assistance or interpreters. For people and households without adequate resources, additional support would ensure income security separate from disability allowances (such as disability pensions, poverty benefits, or other cash transfers). In these cases, any means-tested benefit must account for disability-related costs as described in section 6.1. The support package should be available across the life cycle, considering needs at different ages. Figure 6.1 illustrates the ideal package. Figure 6.2: Package of Support to Persons with Disabilities Depending on the Level of Recognized Needs and Disability Costs Access to certain Access to further Inclusive services support services support services and economic (for example (for example empowerment respite care) personal programs assistance) Disability Type of support Disability pension, poverty Affordable health Concessions to allowance benefit, other care, assistive offset specific (compatible with Top up to income security products + sources of work and other disability (accounting for rehabilitation disability costs schemes) allowance disability costs) All persons Registered persons Persons with significant Persons with high Persons with disabilities with disabilities with disabilities disability needs disability support needs low/no income Source: Original figure for this publication; World Bank and UNICEF 2025. Each policy instrument has distinct advantages and disadvantages that can be assessed across several dimensions: equity, agency, difficulty of implementation, budget requirements, and political support. Equity refers to whether the intervention is likely to reach all eligible people and whether people with greater disadvantages would benefit more or less than other persons with disabilities. Agency concerns the person’s role in the intervention—whether the person is an active or passive recipient. Different interventions involve varying levels of implementation complexity and often require advanced operational systems. Methodological guidelines on assessing household disability-related costs and their implication for participation 63 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Strong equity and agency impacts may appear achievable in theory, but the administrative capacity to deliver them must be considered. Some programs, though administratively convenient, could perform poorly in practice, such as programs that shift responsibilities for providing discounts or services to external actors but without adequate budgets. Table 6.3 outlines the main considerations for each intervention type across these dimensions. Table 6.3: Equity, Agency, and Implementation Considerations in Key Interventions Intervention Equity Agency Implementation Disability allowance Positive; reaches all those Money allocated to person Requires system of registered and likely to be with disability, who can identification and disability progressive (especially if decide how to spend the needs assessment; can differentiate between level allowance provide incentives to of support needs) registration (backbone to the planning of other interventions) Poverty/minimum income Progressive benefit, but Targeted at household level, Complex administration guaranty benefit (factoring often facing take-up issues which does not promote requiring system to assess disability costs) agency of person with household income or disability proxies; requires good systems and skilled staff Provision of special services Good at addressing high Relatively neutral, though May require presence of needs; service accessibility design in access to service specialized and trained potentially an issue (favoring could increase some agency personnel, depending on relatively better-off) aspects type of service Tailored health care package, Equitable and flexible to Relatively neutral (based Complex implementation; concessions on user charges diverse needs; possible on needs) but recognizing assessing entitlements based issues of accessibility individual rights on specific conditions Concessions Potential to offset some Relatively neutral but Vary depending on design disability related costs but recognizing individual rights but necessary to monitor depending on design may budget and impact favor better-off Preferential tax treatments Dependent on design but Need to be informed and Complex, often resulting in tend to support relatively act in relation to specific unclear budget expenditure better-off among persons incentives and complementarity with with disabilities other policies Source: Original table for this publication; World Bank and UNICEF, 2025. A disability allowance that compensates for average costs generally promotes equity (especially if it accounts for functional limitations) and supports agency by allocating directly to the person with disability. Implementing it requires a system that identifies persons with disabilities and assesses needs. The process can help build a disability registry, which supports planning and deliver of other interventions. A poverty benefit that accounts for disability-related costs tends to be progressive but weak on agency, because it targets households, not individuals. It also requires complex eligibility tests, skilled staff, and advanced systems, which often results in take-up issues and no coverage for a percentage of eligible households. Health care packages, no matter how they are financed, can be tailored to the diverse needs of persons with disabilities through cost caps or concessions in user fees. These measures can be equitable and flexible but require careful design, monitoring, and adjustment as needed over time. As with other services, limited accessibility might affect equity. Methodological guidelines on assessing household disability-related costs and their implication for participation 64 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Heavily subsidized or Free specialized services (such as assistive products, personal assistance, interpreters habilitation services) are very important but there are equity concerns to consider when access varies by location, potentially excluding people in remote or poorer areas. These services are sometimes capped or involve user fees, and their equity depends on needs assessment and fee design. (Quasi) universal Disability allowances and registries can improve access and thus equity. Additionally, these services are better suited to address high-needs cases, and agency can increase when persons with disabilities help tailor the service to meet their needs. However, service delivery may requires trained specialists and specific service delivery arrangements. Concessions in other services, such as transportation discounts, are also flexible, but their design needs careful attention to avoid benefiting mostly those who are better off. Effective design and links with other interventions can improve use and increase coverage. Administering concessions requires accounting systems to track budget efforts and beneficiaries. Besides various tax exemptions (which are like concessions), preferential tax treatments such as income tax reductions are likely to favor higher-income people, and even when supporting employment participation, the mechanism tends to lack transparency. Separate cash transfer programs are generally preferred, with defined budgets, clearly identified beneficiaries, and support linked to disability-related costs and combined with other measures. Interventions have different budget implications and levels of political support that determine whether the government will adopt them. A stakeholder analysis can help identify sources of support and opposition to aid in advocacy efforts. Methodological guidelines on assessing household disability-related costs and their implication for participation 65 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix 6.4 Developing a Response Package in Different Contexts Figure 6.1 describes a possible optimum support package to address disability-related costs, but countries vary widely in their progress in developing such interventions. For example, many OECD countries still fail to address all disability costs satisfactorily. Applying findings from disability-related cost studies depends on the country context and requires gradual reform. Analyzing the study’s policy implications can follow three steps: 1. Assess how the study relates to the current official disability status certification system. It may highlight the need to differentiate by disability type, degree of functional difficulties and level of disability support needs, but current data may be lacking. This would help in initiating changes in the disability assessment and disability determination, considering implementation constraints and how to address them. 2. Compare existing programs with the optimum package. Then, cased on the study’s findings and the identified goods and services, categorize the needed changes into three groups: existing programs that need improved coverage or support levels, priority interventions, and interventions for future consideration. 3. Perform preliminary calculations for interventions in the first two groups to evaluate budget needs, equity and agency trade-offs, and implementation feasibility. The analysis can then guide discussions with government analyzing programs and interventions. To illustrate the three-step process, the reminder of this section compares two hypothetical cases in which cash benefits and in-kind services are at different stages of development. These examples are based on typical patterns in various countries and show the type of reforms and policy changes that could unfold over the medium term. Actual reforms will depend on national context and political commitment. Case 1: Limited Package: Small and Partial Support Disability Assessment and Determination Disability assessment and determination involve several stages: initial screening and survey identifies potential cases, followed by assessment by local committees. A medical assessment is conducted only in cases of disagreement or appeal. This process does not extensively assess needs. Although disability status is categorized by impairment type, it does not account for the degree of functional difficulty or identify high support needs. Findings from studies on disability-related cost show that costs vary significantly based on functional difficulties and support needs, yet countries often lack ways to distinguish higher support needs, such as the need for substantial human assistance. Clear criteria are needed to distinguish different levels of support needs. Some countries limit the disability certification criteria to only to persons with severe disabilities, which has drawbacks, because they merge assessment and determination to limit eligibility for cash benefits limiting their abilities to design support package responsive to the diversity of persons with disabilities (Barca et al. 2021). Revising the disability assessment must involve persons with disabilities and their representative organizations to ensure the ability to distinguish disability types and different support needs. Methodological guidelines on assessing household disability-related costs and their implication for participation 66 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Existing Support for Persons with Disabilities Existing support is limited to a flat cash allowance for poor persons with severe disabilities of working age. Some basic health care is available, but it is not tailored to disability-specific needs. Rehabilitation services are minimal, available only in some areas of the country and provided mainly by nongovernmental, faith- based, or disability-related organizations. Access to assistive products is extremely limited and delivered ad hoc, again mostly through nongovernmental actors. Concessions or other support services are nonexistent. Income security relies on employment-based disability insurance that covers only those in public sector and formal jobs. A conditional cash transfer program for poor households with children could potentially reach households with a member with disabilities, but it does not account for the additional disability costs. Possible Expansion of Package in the Medium Term In these settings and based on findings from disability-related cost studies, table 6.4 presents possible actions in three categories: improvement to existing programs, first-priority expansion of programs, and future programs. This outlines a sequence for planning and organizing interventions. Improving current programs could start by expanding disability allowance to children and older persons, replace poverty eligibility assessments by affluence testing and disability pension testing to account for disability-related costs, considering a top up for persons needing substantial personal care. Priority new programs would focus on increasing coverage of community-based care and support and procuring selected assistive devices, considering the needs of different disability types. To improve access to health care and reduce disability-related costs, a system of concessions should be designed to broaden the limited basic health package, which is too narrow for persons with disabilities. These services could expand gradually, which can allow increased coverage of the basic disability allowance and establish a top- up allowance in some cases. Table 6.4: Limited Package: Current Programs, Improvements, and Medium-Term Priorities Potential package Status quo Improvements to Priority new programs Future new components current programs programs All persons with disability Access to basic n.a. Increase coverage of community- Expand 1. Affordable health care, health package; based rehabilitation, centers coverage assistive products plus limited coverage of functioning as disability of services rehabilitation rehabilitation and assessment, provision of and list of assistive products, some services, referral; start assistive 2. Inclusive services and mainly from NGOs procurement of selected assistive devices economic empowerment devices (WHO essential list) programs Registered persons with No concessions n.a. Design and implement n.a. disabilities Concessions to concessions for health care offset specific sources of services disability costs Persons with significant needs Disability allowance Removing Universal 1. Disability allowance only to relatively poverty targeting disability poor working age and replace allowance 2. Access to certain support persons with severe with disability services disabilitiesy pension and affluence tested benefits. Consider differentiating based on degree of functional difficulties: Methodological guidelines on assessing household disability-related costs and their implication for participation 67 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Potential package Status quo Improvements to Priority new programs Future new components current programs programs Persons with very high needs Neither is present n.a. Initiate piloting of community Top-up 1. Top-up to disability based care and support services allowance allowance 2. Access to additional support services Persons with low or no income Contributory Factor extra costs n.a. n.a. Disability pension, poverty disability pension into the eligibility benefit, other income security plus CCT for poor test to CCT households (without program considering extra costs) Source: Original table for this publication; World Bank and UNICEF, 2025. Note: CCT = conditional cash transfer; NGO = nongovernmental organization; WHO = World Health Organization. Advanced Package Disability Assessment and Determination The disability assessment focuses primarily on impairment and a medical assessment. The system distinguishes disability severity types and identifies three groups with different degrees of severity of impairment. While the system already distinguishes between groups and does not group all persons with disabilities uniformly. Reform is underway to broaden the assessment of support needs in different domains across the life cycle to align with recommendations of the Committee on the Rights of People with Disabilities, persons with disabilities. The study on disability-related costs which confirmed that costs increase with the level of support needs and differ in structure by age, disability type can inform the reform the existing disability assessment and determination mechanism. Existing Support for Persons with Disabilities In this scenario, the country already provides a relatively complex support package for persons with disabilities. A universal health package includes habilitation and rehabilitation services, some assistive devices, inclusive education with support teachers, and dedicated employment services. However, user fees and copayments apply in many health services, but people with recognized disability status receive concessions. Public transportation concessions are available but are effective only in large cities. All people with recognized disability status receive a universal disability allowance, with amounts varying by disability group to reflect differing needs. Day-care services are also available but do not provide pathways to inclusion. The mainstream poverty benefit adjusts the eligibility assessment to account for disability- related costs and is compatible with disability allowance. Possible Expansion of Package in the Next Five Years Findings from the disability-related costs assessments should guide a review of the current systems of concessions for user fees in health care. The system should respond better to the needs of different disability groups and types. The adequacy of the disability allowance by group should be verified, and the poverty benefit should match the estimated disability-related expenditures. Two main service expansion initiatives should be considered: developing home based care and personal assistance services and reforming day-care services, and broadening the list of assistive products and the type of services covered under the health package. These changes should be based on the items identified in the exploratory GSR and vary by disability type. Methodological guidelines on assessing household disability-related costs and their implication for participation 68 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Additional programs to consider later include transportation services such as transportation to schools, extending public transportation concessions beyond large cities, expanding health care services and assistive products; and providing respite care services. Table 6.5 summarizes this review. Table 6.5: Advanced Package: Current Programs, Improvements, and Medium-Term Priorities Potential package Status quo Improvements to Priority new programs Future new programs components current programs All persons with Access to health n.a. Increase in the health Expand services and disability 1. package, including package based on the list of assistive devices Affordable health care, some rehabilitation list of rehabilitation Transportation to assistive products plus services and assistive services and assistive school for specific rehabilitation products devices identified in disability types the study (varies by 2. Inclusive services Some inclusive disability type) and economic services in education empowerment and employment programs services Registered persons System of concessions Review system of n.a. with disabilities to reduce health user concessions in relation Concessions to offset fees plus concession in to disability groups specific sources of public transportation and type of disability, disability costs especially for people living in rural areas Persons with significant Universal disability Verify adequacy of n.a. needs allowance allowance 1. Disability allowance Day-care services Reform day care 2. Access to certain to favor inclusive support services education and employment Persons with very high The disability Verify adequacy of Home-based care, Respite care services needs 1. allowance differentiate allowance personal assistant, Top-up to disability amount based on the respite care Reform day care allowance disability groups to favor inclusive 2. Access to additional Day care services education and support services employment Persons with low Poverty benefit Review extra costs n.a. n.a. or no income (considering extra assessment to Disability pension, costs) affect both eligibility poverty benefit, other assessment and income security amount of benefit Sources: Original table for this publication; World Bank and UNICEF 2025. Note: CCT = conditional cash transfer; NGO = nongovernmental organization; WHO = World Health Organization. Georgia was the first country to implement an exploratory GSR focused on children with disabilities and based on an earlier map of existing benefits and services. It also designed a revised disability assessment for children that considers diverse needs (medical, functional, and social). Box 6.3 details how Georgia is developing a response package for children with disabilities. Methodological guidelines on assessing household disability-related costs and their implication for participation 69 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Box 6.3:   Developing a Response Package in Georgia Georgia is working toward a more equitable system by modifying support measures to better reflect the needs of children with disabilities. Although individual packages were discussed, the approach was considered premature because of the discretion it would require from front-line workers. Instead, the current approach focuses on developing a set of response packages tailored to different groups of children with disabilities. For example, the basic support package for children with physical disabilities is considering the following elements: a uniform cash benefit for all children with recognized disability status; an expanded list of assistive technologies and devices; specialized services such as early intervention, habilitation and rehabilitation program, recreation services at medical resorts, and targeted therapy sessions; personal assistance and care, including day-care services, home-based care, personal assistants, and respite services for parents; a pre-school and school inclusion package covering transportation to school, a catch-up program with support teacher, and an assistant in school; and specialized health services, such as hygiene products and sedation for dental care. Table B6.3.1 shows how these components have been assigned to different groups of children with physical disabilities. Among the assistive devices, some have been included in the pilot list and others discarded. Specialized therapies such as aqua therapy, art and music, and horse-riding therapy were not included. Respite services for parents, transportation to school, and hygiene products were selected for pilot implementation and gradual inclusion in the response package. Table B6.3.1. Example: Developing a Package for Children with Physical Disabilities in Georgia Priority groups Package components Group 1: current entitlement Group 2: to be piloted Group 3: excluded for now Addressing average Cash benefit disability-related costs Assessment of disadvantage Manual wheelchair, Stander (with postural Adapted chair and table, car wheelchair with postural control), walker, walker with seat, development toys and support, electric wheelchair, postural control, transfer materials, hammock or swing adapted toilet seat, bath crane or hoist, transfer or trampoline, adapted cup or shower chair, orthoses board, portable ramp, presser or bottle, cutlery (splints), diapers, bathroom relief cushion, orthopedic handrails or grab bars mattress, orthopedic shoes, pillow for head Specialized services Early intervention service, Aqua therapy and couch; art habilitation or rehabilitation therapy, ride therapy, music program, recreation service therapy, or hydromassage at a medical resort Personal assistance and Day-care service (regular and Respite service for parents care specialized), home-based care, personal assistant (regular, specialized) Preschool and school Catch-up program and Transportation to school inclusion package special teacher; Third person assistant at the school Specialized health care Sedation for dental services Hygiene products package Source: Original table for this publication; World Bank and UNICEF 2025. Methodological guidelines on assessing household disability-related costs and their implication for participation 70 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix References Asuman, Derek, Charles Godfred Ackah, and Frank Agyire-Tettey. 2021. “Disability and Household Welfare in Ghana: Costs and Correlates.” Journal of Family and Economic Issues 42 (4): 633–49. https://doi. org/10.1007/s10834-020-09741-5. Atkinson, Anthony B. 2019. Measuring Poverty around the World. Princeton, NJ: Princeton University Press. https://doi.org/10.2307/j.ctvc77fd6. Barca, Valentina, Madhumitha Hebbar, Alexandre Cote, Emrys Schoemaker, Susan Enfield, Rebecca Holmes, Emily Wylde, Ludovico Carraro, Paul Heinrich Horsters, Dominique Leska, et al. 2021. “Inclusive Information Systems for Social Protection: Intentionally Integrating Gender and Disability.” London; Bonn, Germany; and Barton, Australia: UK Foreign, Commonwealth, and Development Office; Deutsche Gesellschaft für Internationale Zusammenarbeit; and Australian Department of Foreign Affairs and Trade. https://www.social-protection.org/gimi/gess/RessourcePDF.action;jsessionid=NfSS_ Uu9Z3e8ydkMR0nnOKG0YGCH2SOVke7YalLreNILt5zGLoHP!1818001608?id=57600. Berthoud, Richard, Jane Lakey, and Stephen McKay. 1993. The Economic Problems of Disabled People. London: Policy Studies Institute. Bradshaw, Jonathan, ed. 1993. Budget Standards for the United Kingdom. Avebury, UK: Avebury. Bradshaw, Jonathan, Sue Middleton, Abigail Davies, Nina Oldfield, Noel Smith, Linda Cusworth, and Julie Williams. 2008. A Minimum Income Standard for Britain: What People Think. York, UK: Joseph Rowntree Foundation. https://www.jrf.org.uk/report/minimum-income-standard-britain-what-people-think. Carraro, Ludovico. 2024. “Testing Poverty Lines for Unaccounted Needs of People with Disabilities: An Application to Bosnia and Herzegovina.” Paper presented at the 22nd European Network for Social Policy Analysis Annual Conference, “Public Values of the Welfare State to Manage Sustainable Welfare and Well- Being,” Tampere, Finland, August 28–30. https://espanet2024.exordo.com/programme/presentation/463. Carraro, Ludovico, Alex Robinson, Rosela Agcaoili, Bonn Michael Canoza, Airabelle Leyda, and Zoe Dominique Cunanan. 2024. “Are Concessions and Discounts for Children with Disabilities Supporting Social Inclusion and Participation?” In Social Welfare and Development Journal CY 2023, Philippines Policy Development and Planning Bureau, 100–120. Manila: government of the Philippines. https:/ /pdpb.dswd. gov.ph/2024/01/social-welfare-and-development-journal-cy-2023/. Carraro, Ludovico, Alex Robinson, Bilal Hakeem, Abner Manlapaz, and Rosela Agcaoili. 2023. “Disability- Related Costs of Children with Disabilities in the Philippines.” International Journal of Environmental Research and Public Health 20 (13): 6304. https://doi.org/10.3390/ijerph20136304. CIP (Center for Inclusive Policy). 2024. The Direct Costs of Disability to Families in Tamil Nadu: A Study Report on the Goods and Services Required by Persons with Disabilities in Tamil Nadu for Full and Effective Participation. Washington, DC: CIP. https://inclusive-policy.org/wp-content/uploads/2024/06/Direct- Costs-of-Disability-to-Families-in-Tamil-Nadu_June-2024.pdf. Cremin, Kevin. 2018. “Adequate Standard of Living and Social Protection.” In The UN Convention on the Rights of Persons with Disabilities: A Commentary, edited by Ilias Bantekas, Michael Ashley Stein, and Dimitris Anastasiou, 808–33. Oxford, UK: Oxford University Press. https:/ /doi.org/10.1093/ law/9780198810667.003.0029. Cullinan, John, Brenda Gannon, and Sean Lyons. 2011. “Estimating the Extra Cost of Living for People with Disabilities.” Health Economics 20 (5): 582–99. https://doi.org/10.1002/hec.1619. Methodological guidelines on assessing household disability-related costs and their implication for participation 71 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Davis, Abigail, Donald Hirsch, and Matt Padley. 2018. “The Minimum Income Standard as a Benchmark of a ‘Participatory Social Minimum.’” Journal of Poverty and Social Justice 26 (1): 19–34. https://doi.org/10.133 2/175982717X15087736009278. Deaton, Angus, and Salman Zaidi. 2002. “Guidelines for Constructing Consumption Aggregates For Welfare Analysis.” Living Standards Measurement Study Working Paper 135, World Bank, Washington, DC. https:/ /documents.worldbank.org/en/publication/documents-reports/ documentdetail/206561468781153320/Guidelines-for-constructing-consumption-aggregates-for- welfare-analysis. Deeming, Christopher. 2020a. “An Introduction to Minimum Income Standards and Reference Budgets: International and Comparative Policy Perspectives.” In Minimum Income Standards and Reference Budgets: International and Comparative Policy Perspectives, edited by Christopher Deeming, 3–24. Bristol, UK: Bristol University Press. https://doi.org/10.46692/9781447352976.003. Deeming, Christopher, ed. 2020b. Minimum Income Standards and Reference Budgets: International and Comparative Policy Perspectives. Bristol, UK: Bristol University Press. https://doi.org/10.2307/j.ctv125jsbv. Dobson, Barbara, and Sue Middleton. 1998. Paying to Care: The Cost of Childhood Disability. York, UK: Joseph Rowntree Foundation. Gijbels, Eef, Julie Vinck, and Wim Van Lancker. 2023. Harmonize Financial Support for Children with Care Needs. Leuven, Belgium: Center for Sociological Research, Leuven University. https:/ /cdn.nimbu. io/s/5s8z9pq/channelentries/yfz8tgy/files/1701682675517/2023_09-1_rapport_19-1_swvg_mjp_14_ rapport-zorgbehoefte-ondersteunen_deel_1.pdf?le8a07a (in Dutch). Goedhart, Theo, Victor Halberstadt, Arie Kapteyn, and Bernard van Praag. 1977. “The Poverty Line: Concept and Measurement.” The Journal of Human Resources 12 (4): 503–20. https://doi. org/10.2307/145372. González, Paula. 2020. “On the Design of Equity-Oriented Pharmaceutical Copayments.” SERIEs 11 (2020): 179–202. https:/ /doi.org/10.1007/s13209-019-00207-y. Groce, Nora E., and Daniel Mont. 2017. “Counting Disability: Emerging Consensus on the Washington Group Questionnaire.” The Lancet Global Health 5 (7): e649–50. https://doi.org/10.1016/S2214- 109X(17)30207-3. Henry, Gary T. 1990. Practical Sampling. Newbury Park, CA: SAGE Publications, Inc. https://doi. org/10.4135/9781412985451. Hirsch, Donald, and Katherine Hill. 2016. “The Additional Cost of Disability: A New Measure and Its Application to Sensory Impairment.” Disability and Society 31 (7): 897–913. https://doi.org/10.1080/09687 599.2016.1221334. ILO (International Labour Organization). 2024. World Social protection Report 2024–26: Universal Social Protection for Climate Action and a Just Transition. Geneva: International Labour Organization. https://www. ilo.org/sites/default/files/2024-09/WSPR_2024_EN_WEB_1.pdf. Indecon (Indecon International Research Economists). 2021. The Cost of Disability in Ireland. Dublin: Department of Social Protection, government of Ireland. https://www.gov.ie/en/publication/1d84e-the- cost-of-disability-in-ireland-research-report/. Methodological guidelines on assessing household disability-related costs and their implication for participation 72 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Jones, Andrew, and Owen O’Donnell. 1995. “Equivalence Scales and the Costs of Disability.” Journal of Public Economics 56 (2): 273–89. https://doi.org/10.1016/0047-2727(93)01416-8. Kapteyn, Arie, Peter Kooreman, and Rob Willemse. 1988. “Some Methodological Issues in the Implementation of Subjective Poverty Definitions.” The Journal of Human Resources 23 (2): 222–42. https:// doi.org/10.2307/145777. MacMahon, Bernadette, Hannah Boylan, and Robert Thornton. 2022. “Care at Home: Costs of Care Arising from Disability.” Tullamore, Ireland: Family Carers Ireland. https://www.budgeting.ie/download/pdf/ care-at-home-costs-of-care-arising-from-disability-2022.pdf. MacMahon, Bernadette, and Robert Thornton. 2020. “Minimum Essential Standards of Living Research in Ireland.” In Minimum Income Standards and Reference Budgets: International and Comparative Policy Perspectives, edited by Christopher Deeming, 39–54. Bristol, UK: Bristol University Press. https://doi. org/10.46692/9781447352976.005. Mancini, Giulia, and Giovanni Vecchi. 2022. On the Construction of a Consumption Aggregate for Inequality and Poverty Analysis. Washington, DC: World Bank. http://documents.worldbank.org/curated/ en/099225003092220001/P1694340e80f9a00a09b20042de5a9cd47e. Middleton, Sue, Karl Ashworth, and Robert Walker. 1994. Family Fortunes: Pressures on Parents and Children in the 1990s. London: Child Poverty Action Group. Middleton, Sue. 2000. “Agreeing Poverty Lines: The Development of Consensual Budget Standards Methodology.” In Researching Poverty, by Jonathan Bradshaw and Roy Sainsbury, 59–76. Oxfordshire, UK: Routledge Publishing. https://doi.org/10.4324/9781315183336. Mitra, Sophie, Michael Palmer, Hoolda Kim, Daniel Mont, and Nora Groce. 2017. “Extra Costs of Living with a Disability: A Review and Agenda for Research.” Disability and Health Journal 10 (4): 475–84. https:// doi.org/10.1016/j.dhjo.2017.04.007. Mont, Daniel, Lena Morgon Banks, Ludovico Carraro, Alex Cote, Jill Hanass-Hancock, Sophie Mitra, Zachary Morris, Mercoledi Nasiir, and Monica Pinilla-Roncancio. 2023. “Methods for Estimating the Impact of Disability Costs for Designing Inclusive Policies.” Disabilities 3 (4): 539–49. https://doi. org/10.3390/disabilities3040034. Morciano, Marcello, Ruth Hancock, and Stephen Pudney. 2015. “Disability Costs and Equivalence Scales in the Older Population in Great Britain.” Review of Income and Wealth 61 (3): 494–514. https://doi. org/10.1111/roiw.12108. Morris, Zachary, Stephanie Rennane, Nanette Goodman, Daniel Mont, Michelle Ballan, Amber Davis, Stephen McGarity, Brooke Ellison, and Ghenet Weldeslassie. 2024. “The Disability Squeeze: Out-of- Pocket Expenses and Unmet Needs for Disability-Related Goods and Services in the US.” RAND Working Paper Series. https://doi.org/10.2139/ssrn.5019619. NDSD (National Department of Social Development). 2015. Elements of the Financial and Economic Costs of Disability to Households in South Africa. Johannesburg: National Department of Social Development, Republic of South Africa. Padley, Matt, and Abigail Davis. 2020. “Applying the Minimum Income Standard in Diverse National Contexts.” In Minimum Income Standards and Reference Budgets: International and Comparative Policy Perspectives, edited by Christopher Deeming, 241–54. Bristol, UK: Bristol University Press. https://doi. org/10.46692/9781447352976.019. Methodological guidelines on assessing household disability-related costs and their implication for participation 73 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Palmer, Michael, Jenny Williams, and Barbara McPake. 2019. “Standard of Living and Disability in Cambodia.” The Journal of Development Studies 55 (11): 2382–2402. https://doi.org/10.1080/00220388.2 018.1528349. Penne, Tess, Ilse Cornelis, and Berenice Storms. 2020. ‘All We Need Is…: Reference Budgets as an EU Policy Indicator to Assess the Adequacy of Minimum Income Protection’. Social Indicators Research 147 (February). https://doi.org/10.1007/s11205-019-02181-1. Pinilla-Roncancio, Mónica. 2023. “Multidimensional Measures and the Extra Costs of Disability: How Are They Related?” International Journal of Environmental Research and Public Health 20 (3): 2729. https://doi. org/10.3390/ijerph20032729. Pollak, Robert A., and Terence J. Wales. 1979. “Welfare Comparisons and Equivalence Scales.” The American Economic Review 69 (2): 216–21. https:/ /www.jstor.org/stable/1801646. Ravallion, Martin. 1998. “Poverty Lines in Theory and Practice.” Living Standards Measurement Study Working Paper 133, World Bank, Washington, DC. https:/ /documents1.worldbank.org/curated/ en/916871468766156239/pdf/multi-page.pdf. Ravallion, Martin, and Michael Lokshin. 2006. “Testing Poverty Lines.” Review of Income and Wealth 52 (3): 399–421. https:/ /doi.org/10.1111/j.1475-4991.2006.00196.x. Smith, Noel, Sue Middleton, Kate Ashton-Brooks, Lynne Cox, Barbara Dobson, and Lorna Reith. 2004. “Disabled People’s Costs of Living: More than You Would Think.” York, UK: Joseph Rowntree Foundation. https://www.jrf.org.uk/report/disabled-peoples-costs-living. UNICEF 2025. “Personas con discapacidad en el Perú: Costos que enfrentan para participar en igualdad de condiciones.” Lima: UNICEF. https://www.unicef.org/peru/informes/personas-con-discapacidad-en-el- per%C3%BA. Townsend, Peter. 1979. Poverty in the United Kingdom. London: Allen Lane and Penguin Books. https:// www.poverty.ac.uk/free-resources-books/poverty-united-kingdom. United Nations. 2024. “Disability and Development Report 2024: Accelerating the Realization of the Sustainable Development Goals by, for and with Persons with Disabilities.” United Nations Publications. New York. UNICEF (United Nations Children’s Fund). 2023. “The Cost of Raising a Child with Disabilities in Georgia: The Goods and Services Required for the Equal Participation of Children with Disabilities.” New York: UNICEF. UNICEF (United Nations Children’s Fund) Philippines. 2022. “Cost of Raising Children with Disabilities in the Philippines.” Manila: UNICEF Philippines. Walker, Robert. 1987. “Consensual Approaches to the Definition of Poverty: Toward an Alternative Methodology.” Journal of Social Policy 16 (2): 213–26. https://doi.org/10.1017/S0047279400020389. Wilkinson-Meyers, Laura, Paul Brown, Robert McNeill, Philip Patston, Sacha Dylan, and Ronelle Baker. 2010. “Estimating the Additional Cost of Disability: Beyond Budget Standards.” Social Science and Medicine 71 (10): 1882–89. https://doi.org/10.1016/j.socscimed.2010.08.019. Zaidi, Asghar, and Tania Burchardt. 2005. “Comparing Incomes When Needs Differ: Equivalization for the Extra Costs of Disability in the UK.” Review of Income and Wealth 51 (1): 89–114. https://doi.org/10.1111/ j.1475-4991.2005.00146.x. Methodological guidelines on assessing household disability-related costs and their implication for participation 74 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Appendix A. Methodologies Used to Estimate Disability-Related Costs Assessing disability-related costs involves understanding the specific needs of persons with disabilities and their families. The methods used for this draw from broader approaches to assess needs across the whole population. The most common approaches are found within the budget standards methodology, including nutrition-based methods, expert-driven reference budgets, and consensual budget standards. Four main approaches are used to assess disability-related costs: ö Consumption patterns analysis (CPA) and standard of living method (SOL), which use existing nationally representative household budget surveys and can often be combined. ö Goods and services used (GSU), which relies on household surveys covering only persons with disabilities and their families. ö Goods and services required (GSR), which gathers input from experts and focus groups. Each method has strengths and limitations related to their partly different objectives, and the choice of method or combination of methods depends on the country context. ö Nationally survey analysis estimates average disability-related expenses, assesses their impact on poverty, and also assesses the effectiveness of programs serving persons with disabilities. However, it underestimates costs in contexts in which disability expenditures are low because needs are largely unmet. ö Goods and services used analysis measures policy impact across disability types and estimates what people spend on disability-related goods and services. However, it tends to underestimate disability- related expenses, because it does not account for people’s income and living standards. ö Goods and services required method tries to look beyond what people spend and account for what they need. It does not aim to produce national cost estimates but illustrates the range and composition of costs policy areas and the diverse needs of different disability types. Methods to measure disability-related costs try to assess the needs of persons with disabilities to achieve a certain living standard. These methods are based on general approaches to measuring budget standards for the overall population. The objective in budget standards is to understand which goods and services will meet the needs of persons with disabilities and enable them to participate equally in society. Methodological guidelines on assessing household disability-related costs and their implication for participation 75 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Budget Standards and Household Needs Assessment The budget standards approach defines the basket of goods and services that provides people with an adequate and decent living standard in their society.26 The approach has a long history, and the work of British sociologist and social reformer Seebohm Rowntree—a study of poverty in York, England in 1901— is considered the first modern poverty assessment (though previous work started in the 17th century and laid the foundations for the study; see Deeming [2020a] for more information). Identifying household needs and budget standards requires different information sources: normative standards based on expert opinions (for example, what is required for an adequate diet, heating in the house, and so on), information from observed households’ consumption patterns from survey data, and public opinions and consensual agreement on what is required to live adequately in a certain society. Based on the importance of these different sources, three distinct budget standards approaches can be identified: ö Nutrition-based approaches rely mainly on household survey data with detailed consumption expenditure information to define a food basket meeting food-energy requirements and in some cases other nutritional standards. These approaches include the cost of basic needs method (Ravallion 1998), which is used the most in LMIC.27 The basket is priced using observed implicit prices collected in the survey. A nonfood allowance is then added based on the spending patterns of households that spend only enough to meet the food basket cost.28 The seemingly reasonable argument for this approach is that if people do not spend more than the minimum requirement for food, then what they spend for nonfood must be a necessity. ö Expert-driven budgets use expert opinion to identify the basket of goods and services and thus determine the budget components, standards, and expenses. Various normative standards guide the identification of what is needed based on a nutritionally balanced diet but also on nonfood elements, such as utilities, clothing, and so on (see, for example, Bradshaw [1993]). This approach can also use household survey data, especially to determine what is consumed for items that lack specific standards and focus group discussions to validate some of the choices made. ö Consensual budget standards considers people the experts, and they should decide what is required to live in a certain society through a series of deliberative group discussions that require negotiation and consensus on what is needed across different life dimensions. This method also verifies that the food basket is adequate based on certain nutritional standards and can verify and compare the consumption basket’s composition against information from survey data. However, focus group discussions with households are the primary information source. The method originated in professor Robert Walker’s recommendation on the need for informed and negotiated debate among the public (Walker 1987) instead of questions in household surveys (as proposed especially in assessing the subjective minimum income question or survey consensus on necessary items) or simply expert-led approaches. 26. Two other conceptually different methods are the income for social participation and the subjective assessment of minimum income. The income for social participation method determines the income level at which households escape deprivation and achieve a certain degree of social participation. Originally developed by Townsend (1979), this method influenced the relative poverty line used in Europe. However, it does not establish a clear threshold or minimum income but at best identifies a possible range of poverty lines (see Atkinson [2019], 47–49). By contrast, the subjective assessment of minimum income, based on a method first developed in the Netherlands (Goedhart et al. 1977; Kapteyn, Kooreman, and Willemse 1988), estimates the minimum income from an analysis of survey responses to questions about the income required to make ends meet. 27. The cost of basic needs method is different from the basic needs approach developed by, among others, professors Paul Streeten and Frances Stewart. 28. A lower requirement is often calculated considering the nonfood share of those whose total consumption expenditure is equal to the food expenditure basket. However, this usually works only in relatively low-income countries. Methodological guidelines on assessing household disability-related costs and their implication for participation 76 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Figure A.1 compares the three methods, showing the different weight given to the information sources in determining the basket of goods and services needed and its monetary value. Both the expert-driven and consensual standards address the limitations of measuring actual expenditure, because this faces a circular problem: setting adequacy standards by simply observing actual expenditure patterns without an external normative judgement. This issue appears in nutrition-based approaches, but it is very limited. Figure A.1: Different Budget Methods to Determine Needs Nutrition-based Expert-driven Consensual standards Public Experts Survey Experts Experts Public Survey Survey Source: Adapted from Deeming 2020a. All these different methods produce minimum income standards or poverty lines29 that are linked to the attainment of a certain living standard, and they allow setting such standards for households of different size and composition. However, the specific hypothesis of many expert-driven and consensual standards is that all household members are in good health and do not face special needs. Therefore, by default they are clearly inadequate to set standards for households with members with disabilities. Similarly, nutrition- based budget standards do not capture the specific needs of persons with disabilities and testing the validity of the poverty line using welfare ratios and subjective assessments of economic living conditions (Ravallion and Lokshin 2006) can demonstrate whether disability needs have been neglected. For example, a recent application in Bosnia and Herzegovina, where subjective assessments of wellbeing are regressed on the welfare ratio between household consumption expenditure divided by the corresponding poverty line and various demographic characteristics shows that the presence in the household of a member with functional difficulties is associated with a systematic under-estimate of the needs of those households (Carraro 2024). Methods Measuring Disability-Related Costs Because general methods of needs assessment within the budget standards approaches do not account for or purposively abstract from the needs of persons with disabilities, specific methods have been developed to address this gap, even though they represent extensions of those discussed above. Focusing on persons with disabilities presents challenges for both quantitative and qualitative studies. National household surveys are not designed to represent this population, and sampling frames are often lacking. Therefore, while general sampling may include persons with disabilities, whether the sample reflects different disability types is unclear. Qualitative and expert-driven studies, which use nonrepresentative samples, often impose uniform preferences and require generalizations. These limitations affect how needs are assessed and call for caution in interpreting results and examining study design. They also highlight the need to combine methods to address these challenges. 29. Arbitrary poverty lines used for international comparisons, such as the dollar per day poverty lines, or the relative poverty lines are different because no specific standard of living is associated with them as their main purpose is to make comparisons across countries and over time. Methodological guidelines on assessing household disability-related costs and their implication for participation 77 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Four methods can be used to assess disability-related costs: CPA, SOL, GSU, and GSR.30 These methods are discussed in depth in the main report, whereas this appendix provides a analytical assessment of their differences and highlight the specific outputs each method provides. The CPA method examines the distribution of consumption expenditure across different categories (food, housing, clothing, health, transportation, communication, and so on) and identifies major differences between households with and without members with a disability. The comparison between households can be restricted to those with similar demographic composition and similar overall expenditure. However, this analysis alone does not identify disability-related expenditure. It does not use a normative judgment against which to assess the different consumption patterns and focus exclusively on consumption expenditure. Nevertheless, it highlights how the different needs of households that include members with disabilities may result in divergent spending priorities. Using these data, it would also be theoretically possible to compute a poverty line for households with members with disabilities using the cost of basic needs method. Restricting the analysis to such households, it is possible to observe whether for households that meet their food expenditure requirements, the nonfood share is different/higher than that for other households. The problem is that usually there are not enough households with members with disabilities in such surveys to be able to implement this specific calculation. The SOL approach measures the level of income or consumption expenditure associated with reaching a certain living standard, measured using a nonmonetary indicator. It compares such incomes and expenditures for households with and without persons with disabilities, observing that households with a member with disability usually need a higher income or expenditure to reach the same standard of living of other households (this is done controlling for various household characteristics). The difference between the income or consumption expenditure of households with persons with disabilities and other households is taken as the ‘disability-related expenditure’. The method is inferential and, when using income data alone, it cannot identify the source of these disability-related expenditure. However, when using surveys that also have expenditure data, this method can be easily combined with the CPA and explain what type of goods and services are likely to drive the need of higher expenditure. In most LMIC, surveys have expenditure data and so the CPA and the SOL can be considered part of the same analysis. While it is possible to implement such methods using ad-hoc surveys, usually both CPA and SOL use existing nationally representative household surveys, such as household income and expenditure surveys,31 which are used to generate official monetary poverty estimates. Such surveys need to collect a minimum level of information to be able to implement this methodology. More specifically, the survey must (i) measure disability among household members, (ii) collect detailed information on consumption expenditure and (iii) contain nonmonetary indicators of living standards. The GSU method interviews persons with disabilities about their use of goods and services related to their disability, and obtains information on their expenditure, which is considered additional to expenditures incurred by other households. This implicitly asks respondents not only to report what are their expenses but to consider these as additional expenses to what other people need to incur. The method ignores by default the possibility that in some budget areas persons with disabilities might have lower needs (for example, persons with restricted movements might require less food). Nevertheless, by focusing on disability-related items, the instrument can go in much more detail on disability-specific items than standard national household surveys on income and expenditure. While this method can use either a qualitative or quantitative approach, its most common implementation involves quantitative surveys. This method does not make an internal comparison with other households, because it does not collect information about these households with comparable instruments. Therefore, if or when comparisons are made, they rely on external surveys where methodological (sampling and questionnaire design) and 30. See Mitra et al. (2017) and Mont et al. (2023) for an overview of these methods. 31. Other examples of such surveys are household budget surveys, household socioeconomic surveys, or living standards measurement surveys. Methodological guidelines on assessing household disability-related costs and their implication for participation 78 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix implementation differences could affect the results. The GSR approach purposively goes beyond what people spend and use. It tries to determine what is required to reach a certain level of participation or standard of living, with a specific focus on the person with disabilities. Information on what is required is obtained in different ways: asking persons with disabilities about unmet needs, through the contribution of experts, or from a consensual and deliberative group discussions with persons with disabilities and their families. While there are quantitative surveys that try to implement such approach,32 most common applications use qualitative methods. This methodology can include a careful comparison with other households, but this has only been done in few countries, all of them in developed economies. Instead, an approach that is more suitable for LMICLMIC is called an Exploratory GSR, where the study focuses exclusively on persons with disabilities, without a comparison group, and on known disability-related goods and services. While all the three (considering CPA-SOL as one) methods produce some estimates and assessment of disability-related costs, all the methods have advantages and disadvantages. Measures of disability-related costs that only capture expenditure for used products and services can be misleading for four reasons: (i) low incomes of persons with disabilities and their families, (ii) lack of knowledge about disability-specific goods and services, (iii) lack of availability of disability-specific goods and services, and (iv) intra-household discrimination. Because persons with disabilities and their families tend to have low incomes, and lower incomes than the rest of the population, they fail to obtain what they need. Moreover, observed expenditure is also affected by the limited awareness of goods and services that could improve the living standards of persons with disabilities. For example, information about assistive devices is most often lacking in certain contexts. This is then combined with the lack of availability and access to disability-specific goods and services. Intra-household discrimination could also affect actual expenditure, once again limiting the amounts of goods and services acquired for persons with disability within the household. The CPA-SOL method tries to overcome the above constraints by assessing disability-related expenditure conditionally to reaching a certain living standard, and thus measuring different disability-related expenditure depending on the living standards achieved. By observing consumption choices, this method assumes that these represent their preferences, and that people consume what is best for them (revealed preferences). Therefore, this method assumes that people have relevant information about goods and services and that these are available. The strength of this approach is that it shows what is possible in the country context but requires a critical assessment of the assumptions made and their implications. Furthermore, this method partly overcomes the limitation that comes from low incomes, but only if disability-specific goods and services are available and people have some knowledge about them. In countries where there is very low provision of disability-specific items, absent markets and imperfect information, the method is likely to grossly under-estimate what is required. Essentially persons with disabilities and their families do not incur additional costs because needs are not met. Overall, the method heavily depends on survey data analysis and uses as an external judgment of adequacy the ability of reaching specific nonmonetary living standards or (where there are enough observations) food nutritional adequacy (similarly to the nutrition-based general assessments). When the living standards indicators account for unmet needs, the method can also go beyond actual disability-related expenditure and capture elements of the costs of disability, but the monetary quantification of this element is even more indirect. Moreover, the level of detail on the type and structure of disability-related costs that can be produced by this method is very limited. The GSU focuses exclusively on what people use, and although it can collect more detailed information on the type of disability-related expenses and capture the diversity that exists between different disability 32. Quantitative GSR implicitly assumes that persons with disabilities and their families have good knowledge about their needs and existing goods and services that can address such needs. This assumption is especially problematic in developing countries. Methodological guidelines on assessing household disability-related costs and their implication for participation 79 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix types, the methodology risks being biased by not accounting for the households’ income level and their standard of living. The CPA-SOL method relies on revealed preferences, but it also fails to correct for income bias and does not use any external normative judgement. Also, because it is based exclusively on interviews with persons with disabilities, it lacks a comparison with other households. To fill the gaps in what people use and determine what is required, the exploratory GSR method overcomes the constraints of low-incomes, lack of information, lack of access, and risks of intra-household distribution, by involving experts’ judgement. This method has also the advantage of considering the diversity of needs across disability types. However, by assuming preferences there is a risk of introducing differences from what people want even if they have access to information and markets are available. For this reason, all assumptions need to be adequately defended and supported by a clear rationale. The methodology provides only illustrative cases of costs and their structure, but it is unable to produce national-level estimates of these costs with limitations on how the findings can be generalized. It relies only on qualitative assessments. Table A.1 summarizes the differences among these methods across the key dimensions: the main experts in implementing the method, the main data source, the paradigm guiding the method, the standard set to define the costs, the type of costs estimated, whether a counterfactual is usually included in the analysis (a comparison between households with and without a member with disabilities), the validity of findings in terms of external validity (whether findings can be generalized outside the sampled areas and people included) and internal validity (the rigor of the study design in estimating disability-related costs without bias), and main strengths and limitations of the method. The summary of differences across the methodologies does not specify the study outputs and how they Table A.1: Characteristics of the Different Methods on Key Dimensions CPA-SOL GSU GSR Characteristic Existing survey Quantitative survey Qualitative survey Experts Social scientist Social scientist Professionals and the public Data source Household survey Household survey Expert opinions and FGDs Declared and assumed Paradigm Revealed preferences Revealed preferences preferences Standards Adequate living standard Not used Social participation Expenditure required to reach a certain Cost of GSR for social Type of cost Average expenditure of GSU living standard participation Depends on the nature of GSR Counterfactual Present Absent but usually absent in LMIC Both external and internal validity but External validity (with Validity within certain contexts and sampling possible sampling limitations) Only internal validity limitations but biased estimates Assesses presence and relative size of Assesses illustrative costs, but disability-related expenditure but limited Assesses expenditure and Strengths focus is on its structure and information on its structure; enables structure by disability types diversity by disability types correction of poverty estimates Underestimates expenditure in contexts Does not provide national Limitations with poor and underdeveloped services, Underestimates expenditure estimates limiting its applicability Source: Original table for this publication. Note: CPA = consumption pattern analysis; FGD = focus group discussion; GSR = goods and services required; GSU = goods and services used; SOL = standard of living. Methodological guidelines on assessing household disability-related costs and their implication for participation 80 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix can be used, which are presented in table A.2, along the three categories of outputs: map of existing efforts, assessment of disadvantage, and identification of policy areas of intervention. However, each has qualifying aspects. Additionally, a fourth category considers the assessment’s contribution to building political support and raising awareness among the population. For the assessment using CPA-SOL, the level of analysis depends on how many households with members with disabilities are included in the survey (linked directly to the size of the overall sample and to what extent the sample is representative of persons with disabilities), and what relevant information the questionnaire collected: ö How it asks about disability, for example, capturing only the official disability status or by using the Washington Group set of questions capturing the degree of functional limitations (box A.1), ö To what extent the survey covers programs and services for persons with disabilities, ö To what extent the survey captured and identified disability-specific expenses, ö Whether the survey includes relevant modules capturing individual level of expenditure on health and education, and ö Whether the survey contains time use modules or information about caregivers in the household (to capture costs related to human assistance). Additionally, a survey with information on individual access to health, education, and labor participation can provide a broader analysis of differences in outcomes. Box A.1:   Identifying Persons with Disabilities in Household Surveys and Washington Group Questions A consensus supports the use of the Washington Group set of questions in surveys to identify functional difficulties and disability (Groce and Mont 2017, World Bank 2020 ). The short set for adults consists of six questions covering functional difficulties in seeing, hearing, walking, cognition, self-care, and communication. Questions collect a yes or no response and also distinguish the degree of the difficulty (on a scale of none, some difficulty, a lot of difficulty, and cannot do at all). The questions are relatively simple, and survey enumerators can ask them because no medical knowledge is required. The questions provide a proxy for measuring disability that can be introduced in surveys effectively, but doing so does not prevent asking other questions that can identify disability. In countries with official disability status, ask about recognized disabilities only after the questions on functional difficulties. The Washington Group Short Set has several caveats that needs to be considered including possible underestimation of persons with psychosocial disabilities or children with disabilities. Similarly, the extent to which the CPA-SOL helps in identifying potential policy improvements depends on the richness of information in the existing survey, but usually it is confined at providing the scope for adjustment/creation of cash benefits, which could vary based on the degree of functional limitations. Calculations based on the CPA-SOL could support the development of corrections in the assessment of needs of persons with disabilities and their families in means-tested benefits and the adjustment of benefits amounts. The same analysis could provide inputs to the design of allowances for persons with disabilities aimed at compensating for generic disability-related expenses. However, these values would always be based on average needs and therefore should be complemented by more specific in-kind support to address the specific needs of some disability types or when people have high support needs. GSU and GSR even more so can provide more information on the specific list of goods and services needed and the extent of unmet needs—for example, by providing information on relevant assistive devices that government programs currently do not support, or the need of supporting households in accessing occupational therapies sessions, and so on. These assessments could then provide information to complement cash assistance with tailored in-kind support and the development of specific services addressing needs of persons with disabilities. Methodological guidelines on assessing household disability-related costs and their implication for participation 81 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix To fully address the needs of persons with disabilities, a comprehensive set of interventions is needed: cash assistance, support to health care costs (including rehabilitation and assistive devices), targeted subsidies to compensate for specific disadvantages in certain sectors (such as transportation, and utilities), and economic support for labor participation. Cash support helps to address some of the disability-related expenses, but various in-kind measures can address the diverse needs faced across disability types. The different methods have complementarities, and no single method is best, but a choice on the method can be made depending on the study’s objectives and motivations. Table A.3 provides examples of research questions the different methods can answer. Although the different methods have been described as stand-alone studies, combining them is recommended to ensure a more comprehensive assessment. Studies that rely primarily on the analysis of existing household surveys could be complemented by at least some focus group discussions, and those conducting primarily a GSR study could benefit from a background analysis of household survey data. For example, using GSR findings for possible policy development would need to rely on disability prevalence estimates from quantitative studies (either an official registry of persons with disabilities or quantitative survey estimates). The level of effort involved in different methods and a general orientation of the budget required for each are also important considerations. Budget estimates clearly vary depending on the costs of data collection in each country and the complexity of the existing system to be analyzed, so budget estimates should be considered indicative. The CPA-SOL assessment can be done in one to two months for an approximate cost of $15,000 if a relatively recent national household survey exists with the required information. An exploratory GSR study is usually implemented in six months at a cost of at least $50,000. Providing an indicative budget for the GSU is more difficult, because the costs and likely duration of the study increase dramatically when collecting a quantitative random sample is involved. The cost of the quantitative survey depends on the scope and the sampling frame used to conduct it. Assuming that a phone survey is possible (which is unlikely in most countries), the cost could be about $25,000, but other more complex strategies for rigorous sampling would increase the cost substantially. A lack of relevant data from existing household surveys would justify collecting a quantitative sample. Methodological guidelines on assessing household disability-related costs and their implication for participation 82 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Table A.2: Differences in Outputs of Methodologies Used to Assess Disability-Related Costs CPA-SOL GSU Exploratory GSR Outputs Existing survey Quantitative survey Qualitative survey Map of existing efforts Yes Yes Yes Description of programs and services Assessment of effectiveness Yesa Yes Limited Assessment of disadvantage Quantification of costs: Yes (expenditure, Yes, but Yes, illustrated cases if goods and Size underestimated only services exist) Structure of costs Very limited Yes Yes Correction of poverty measures Yes No No Understanding diversity by disability types Limited Yes Yes Quantitative assessment of certain unmet needs Yesa Yes Limited Elements of indirect costs Yesa Partly Limited Burden of unpaid care Yesa Limited Limited Gaps in key outcomes versus rest of populationb Yesa No No Identification of policy areas Possible improvements in cash support Yes, some Yes, some Yes, some Possible improvements in other areas (list of assistive Limited Limited Yes devices, rehabilitation services, and so on) Other dimensions, contributing to Political support Limited Yes Yes Awareness among population Limited Yes Yes Resources and timeline Indicative budget $15,000 $25,000c $50,000 Expected timeframe 2–3 months 4–6 months 6 months Source: Original table for this publication; World Bank and UNICEF 2025. Note: FGD = focus group discussion; GSR = goods and services required; GSU = goods and services used; SOL = standard of living. a. Depending on the relevant information being collected in the household survey used for the analysis b. The assessment of outcomes is additional to the measurement of disability-related costs. c. Assuming that the context permits to conduct a phone survey. Methodological guidelines on assessing household disability-related costs and their implication for participation 83 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix Table A.3: Examples of Research Questions Answered by the Different Methods Questions Methods best suited to answer the questions Current programs for persons with disabilities What existing programs support persons with disabilities? All three methods should provide similar answers to this question. What is the current level of service provision for persons CPA-SOL can answer these questions if the survey collects with disabilities? Are services available? If so, to what extent information about relevant programs for persons with are they accessible and used? Do those who need assistive disabilities (which usually happens only for relatively large technologies receive them? What are the barriers to access? programs). Are there differences in who receives these services? GSU is well suited to answer these questions and could have the sample size needed to answer more precisely across disability types. However, the survey needs to collect information to rank households across income or wealth quintiles to assess any disparity in access across economic wellbeing. GSR cannot answer these questions beyond capturing some general trends. Current expenditure and poverty levels among persons with disabilities and their families What is the direct, out-of-pocket extra cost for households CPA-SOL and GSU are well placed to answer these with members with disabilities? What different expenditures questions. GSU provides more disaggregated details on do persons with disabilities and their families incur? disability-specific expenses and by type of disability, while GSR identifies and assesses only the common type of expenses. How does income differ among persons with and without CPA-SOL is the only method that can answer these disabilities? Are persons with disabilities more likely to live questions. in households with income below the poverty line after accounting for disability-related costs? Can the indirect costs of disability be quantified? To CPA-SOL can answer these questions if relevant questions what extent do family members pass up employment are included. GSU also can address these questions. opportunities, reduce their working hours, or receive lower pay to care for the person with disabilities? Assessment of types of goods and services needed by persons with disabilities What costs and expenses would persons with disabilities Only GSR can address this question. It does not provide a and their families need to incur to achieve equal robust estimate of the cost but provides information on the participation? range of costs and their structure. To what extent are costs and needs different across urban, A combination of different methodologies is needed to rural, and remote areas and the type of disability and age of answer the question fully. GSR can provide information on the person with disability? key drivers of the differences and on needs across disability types. CPA-SOL can answer this question if goods and services are available, but that is limited to aggregated costs, not those specific to the different types of goods and services. Source: Original table for this publication; World Bank and UNICEF 2025. Note: CPA = consumption pattern analysis; GSR = goods and services required; GSU = goods and services used; SOL = standard of living. Methodological guidelines on assessing household disability-related costs and their implication for participation 84 Executive Introduction Consumption Standard of Living Goods and Goods and Measures to address Bibliography Summary Patterns Analysis Method Services Used Services Required disability-related costs & Appendix References Atkinson, Anthony B. 2019. Measuring Poverty around the World. Princeton, NJ: Princeton University Press. https://doi.org/10.2307/j.ctvc77fd6. Bradshaw, Jonathan, ed. 1993. Budget Standards for the United Kingdom. Avebury, UK: Avebury. Carraro, Ludovico. 2024. “Testing Poverty Lines for Unaccounted Needs of People with Disabilities: An Application to Bosnia and Herzegovina.” Paper presented at the 22nd European Network for Social Policy Analysis Annual Conference, “Public Values of the Welfare State to Manage Sustainable Welfare and Well- Being,” Tampere, Finland, August 28–30. https://espanet2024.exordo.com/programme/presentation/463. Deeming, Christopher. 2020a. “An Introduction to Minimum Income Standards and Reference Budgets: International and Comparative Policy Perspectives.” In Minimum Income Standards and Reference Budgets: International and Comparative Policy Perspectives, edited by Christopher Deeming, 3–24. Bristol, UK: Bristol University Press. https://doi.org/10.46692/9781447352976.003. Goedhart, Theo, Victor Halberstadt, Arie Kapteyn, and Bernard van Praag. 1977. “The Poverty Line: Concept and Measurement.” The Journal of Human Resources 12 (4): 503–20. https://doi. org/10.2307/145372. Groce, Nora E., and Daniel Mont. 2017. “Counting Disability: Emerging Consensus on the Washington Group Questionnaire.” The Lancet Global Health 5 (7): e649–50. https://doi.org/10.1016/S2214- 109X(17)30207-3. Kapteyn, Arie, Peter Kooreman, and Rob Willemse. 1988. “Some Methodological Issues in the Implementation of Subjective Poverty Definitions.” The Journal of Human Resources 23 (2): 222–42. https:// doi.org/10.2307/145777. Mitra, Sophie, Michael Palmer, Hoolda Kim, Daniel Mont, and Nora Groce. 2017. “Extra Costs of Living with a Disability: A Review and Agenda for Research.” Disability and Health Journal 10 (4): 475–84. https:// doi.org/10.1016/j.dhjo.2017.04.007. Mont, Daniel, Lena Morgon Banks, Ludovico Carraro, Alex Cote, Jill Hanass-Hancock, Sophie Mitra, Zachary Morris, Mercoledi Nasiir, and Monica Pinilla-Roncancio. 2023. “Methods for Estimating the Impact of Disability Costs for Designing Inclusive Policies.” Disabilities 3 (4): 539–49. https://doi. org/10.3390/disabilities3040034. Ravallion, Martin. 1998. “Poverty Lines in Theory and Practice.” Living Standards Measurement Study Working Paper 133, World Bank, Washington, DC. Ravallion, Martin, and Michael Lokshin. 2006. “Testing Poverty Lines.” Review of Income and Wealth 52 (3): 399–421. https:/ /doi.org/10.1111/j.1475-4991.2006.00196.x. Townsend, Peter. 1979. Poverty in the United Kingdom. London: Allen Lane and Penguin Books. https:// www.poverty.ac.uk/free-resources-books/poverty-united-kingdom. Walker, Robert. 1987. “Consensual Approaches to the Definition of Poverty: Toward an Alternative Methodology.” Journal of Social Policy 16 (2): 213–26. https://doi.org/10.1017/S0047279400020389. Methodological guidelines on assessing household disability-related costs and their implication for participation 85