S URV EY BRIEF Disaggregated Survey Data on Taxation to Improve Policy Design A PERSPECTIVE FROM THE ETHIOPIA SOCIOECONOMIC SURVEY (2018/19) Alemayehu A. Ambel, Hitomi Komatsu, Gayatri Koolwal, Asmelash H. Tsegay and Manex B. Yonis1 June 2021 A disaggregated analysis of the tax burdens and economic needs of the most economically vulnerable - such as poor women and men, informal workers, and owners of micro- and small enterprises - is crucial for designing equitable and well-targeted tax and public spending policies. This is particularly important in low-income countries, where formal and informal tax systems often exist in parallel, and administrative data is sparse. Availability of data on tax payments by households and non-farm enterprises, individual-level employment and asset ownership, and contributions to community institutions and infrastructure programs can reveal important distributional implications for tax policy design going forward. This survey brief summarizes findings from the tax module of the nationally representative 2018/19 Ethiopia Socioeconomic Survey (ESS), discussed in detail in a new report, The Taxation of Individuals, Households, and Nonfarm Enterprises in Ethiopia: Findings from the 2018/19 Ethiopia Socioeconomic Survey. The tax module covers different types of formal and informal taxes that households, businesses, and individuals pay, as well as informal contributions towards services and infrastructural improvements in the community. Findings from the multi-topic survey also reveal different tax burdens by socioeconomic and demographic groups, as well as across rural and urban areas, that are important for national tax policy design and targeting.1 1 Affiliations: DEC Data Group, World Bank Acknowledgements: The fourth wave of the Ethiopia Socioeconomic Survey was conducted in 2018/2019 and a collaborative project of the Ethiopia Central Statistics Agency (CSA) and the World Bank with financial support from the Bill and Melinda Gates Foundation as part of the LSMS-ISA and LSMS+ programs. The tax module was added in the ESS 2018/19, with financial support from FCDO. Disclaimer: The findings, interpretations, and conclusions are entirely those of the authors. They do not necessarily represent the views of the World Bank Group, its Executive Directors, or the governments they represent. 1 Disaggregated Survey Data on Taxation to Improve Policy Design A PERSPECTIVE FROM THE ETHIOPIA SOCIOECONOMIC SURVEY (2018/19) S U R VE Y B R I E F FINDINGS ON Figure 1 INDIVIDUAL, HOUSEHOLD Annual Business Income Tax Paid (Birr), AND BUSINESS TAXATION by Sex of Owner 3500 BUSINESS TAXES 3000 Respondents provided detailed information 2500 on household business activity during the 2000 12 months preceding the survey. Twenty 1500 three percent of households (about 1000 38 percent in urban and 16 percent in rural 500 areas) have at least one household business. 0 Most of these businesses are micro ALL FIRMS MEN WOMEN OWNERS OWNERS enterprises, in small scale activities in retail, Mean Median transport, home-based shops and services, with 97 percent employing no more than five Source: World Bank (2021). The Taxation of Individuals, Households, and Nonfarm Enterprises in employees, and about 85 percent of which Ethiopia: Findings from the 2018/19 Ethiopia Socioeconomic Survey. are own-account (with no employees). About 12 percent of household non-farm EMPLOYMENT INCOME: enterprises (18 percent in urban and IMPUTED TAXES 6 percent in rural areas) pay the business Among employees, the survey collected income tax. Female business owners detailed information on income earned by pay nearly 25 percent less in business wage and salary workers. The estimated income tax compared to male business taxes paid on this income can be imputed owners (Figure 1). Twenty one percent using the rate schedule in the employment of household enterprises had paid the income tax (EIT) legislation. Based on gross business license fee, with the majority of salary and wage data collected in the survey, the tax-paying group in urban areas. Only about 72 percent of employees earned 6 percent of enterprises paid the municipal income above the EIT exemption threshold service fees. Further analysis could shed of 600 Birr/month. Seventy five percent of light on the reasons for the gender male earners are in the tax net compared to differences in tax payments, including 67 percent of female earners, because male differences by business size and industry. earnings are higher. Figure 2 shows that the gender differences in wage earnings are much more pronounced in urban areas. 2 Disaggregated Survey Data on Taxation to Improve Policy Design A PERSPECTIVE FROM THE ETHIOPIA SOCIOECONOMIC SURVEY (2018/19) S U R VE Y B R I E F 57 percent of whom paid the land use fee and Figure 2 the housing tax. Male- and female-headed Taxable Income Distribution by Place of Residence and Sex households paid about the same amounts in taxes. In general, however, property valuation URBAN AREAS in urban areas such as Addis is outdated and .000025 does not reflect changes in urbanization. .00002 INFORMAL TAXES AND DENSITY .000015 CONTRIBUTIONS .00001 In Ethiopia, households are more likely to interact with local formal and informal systems 5.000e-06 of taxation. Informal taxes include in-kind and 0 cash contributions by households to finance 0 100 000 200 000 300 000 public goods, such as roads and schools; ANNUAL WAGE (IN BIRR) and informal taxation by local public officials, Men Women enforced through social channels (community fundraisers and direct collection; also see RURAL AREAS a discussion by Olken and Singhal, 2011). 2 .00003 Existing questions in the survey were refined to differentiate between the various informal cash and in-kind labor contributions made to .00002 DENSITY local state and non-state actors. The findings reflect the significant contributions made, .00001 particularly by rural households (Figure 3). Nationally, about 44 percent of households 0 contribute to informal social security 0 100 000 200 000 300 000 institutions and 40 percent to religious ANNUAL WAGE (IN BIRR) institutions; in rural areas, these shares go Men Women up to 49 and 44 percent, respectively. Twenty two percent of households contribute to Source: World Bank (2021). local community development activities and The Taxation of Individuals, Households, and Nonfarm Enterprises in Ethiopia: Findings from the 2018/19 Ethiopia Socioeconomic Survey. 15.2 percent to social and political activities, with higher shares again in rural areas. Twelve percent of individuals in the sample make free labor contributions to social and local PROPERTY TAXES (URBAN AREAS) development activities. The survey asked urban households reporting ownership of a dwelling on whether they paid an urban land use fee and housing 2 Informal taxation can have one advantage: local officials tax. About 42 percent of urban households better understand both their community’s needs and the distribution of income. For additional recent studies on in- lived in privately-owned dwellings, about formal taxation, also see Khan et al. 2016; and Walker 2018. 3 Disaggregated Survey Data on Taxation to Improve Policy Design A PERSPECTIVE FROM THE ETHIOPIA SOCIOECONOMIC SURVEY (2018/19) S U R VE Y B R I E F AGRICULTURAL INCOME TAX AND Figure 3 LAND USE (RURAL AREAS) Share of households making informal local contributions in Ethiopia In 2019, about 79 percent of the population 60 lived in rural areas, and agriculture (farming 50 or livestock) is practiced by 96 percent of rural 40 households. About three-quarters of rural 30 households pay some type of agricultural tax. Rural farmers are required to pay a rural land 20 use fee and an agricultural income tax, the 10 total of which usually takes the form of a “crop 0 Social and political Community development Religious institutions Informal Free labor social security contribution farming income tax.” About 74 percent of activities activities institutions (individuals) rural households pay this tax. Pastoralists are National Urban Rural subject to a livestock income tax according to Source: World Bank (2021). the number of livestock they own, but is paid The Taxation of Individuals, Households, and Nonfarm Enterprises in Ethiopia: Findings from the 2018/19 Ethiopia Socioeconomic Survey. by very few households (2 percent). Each region sets their own tax rates. The rates are typically determined according to the amount of agricultural land. For example, There are also gender differences when in Oromia, tax liabilities also vary between examining the share of households making whether farmers have access to irrigation. informal local contributions. While male- Consumption-poor households tend to have headed households are more likely to make a higher tax burden, and separate research these contributions overall, female-headed shows that female adult-only households have households that contribute to informal social a higher burden (Komatsu et al., 2021). security institutions pay about the same amount as male-headed households. Among female-headed households, contributions to informal social security institutions also tend to be highest, whereas for men, the greatest average contributions are towards religious institutions. In future analyses, the data allows to make a distinction between the different types of tax and non-tax payments and develop a more comprehensive picture of the overall tax burdens of individuals and households. 4 Disaggregated Survey Data on Taxation to Improve Policy Design A PERSPECTIVE FROM THE ETHIOPIA SOCIOECONOMIC SURVEY (2018/19) S U R VE Y B R I E F IMPLICATIONS OF help us better understand the impact of DISAGGREGATED changes in tax policy and tax administration. DATA COLLECTION The responses to these questions will FOR POLICY DESIGN enable researchers and policymakers to develop a more detailed understanding The Ethiopia Socioeconomic Survey provides of the prevalence and magnitude of local new and disaggregated data on taxation and informal taxes, as well as a more of individuals, households, and household comprehensive picture of the overall tax businesses, and is an important step toward burden faced by individuals and households. collecting micro-survey data on taxation. Data disaggregated at these levels, alongside The survey findings show how improving the administrative data that can be used in evidence base on taxation at the subnational assessments of government programs and level can raise relevant issues for policy initiatives, is important for shaping policy design, and help enhance government targeting and vital for a well-functioning and capacity to develop better-targeted and more well-informed tax collection system. equitable policies on taxation and spending. The 2018-19 Ethiopia Socioeconomic Survey Microeconomic survey data collection can can be accessed here. also complement administrative data to References Khan, Adnan, Asim Khwaja, and Benjamin Olken. 2016. “Tax Farming Redux: Experimental Evidence on Performance Pay for Tax Collectors,” The Quarterly Journal of Economics 131 (1): 219–71. Komatsu, Hitomi, Alemayehu Ambel, Gayatri Koolwal, and Manex Boule Yonis. 2021. “Gender and Tax Incidence of Rural Land Use Fee and Agricultural Income Tax in Ethiopia.” World Bank Policy Research Working Paper 9715, World Bank. Olken, Ben, and Monica Singhal. 2011. “Informal Taxation.” American Economic Journal: Applied Economics 3(4): 1–28. Walker, Michael. 2018. “Informal Taxation and Cash Transfers: Experimental Evidence from Kenya.” Working Paper C-89334-KEN-1, International Growth Centre. World Bank (2021). The Taxation of Individuals, Households, and Nonfarm Enterprises in Ethiopia: Findings from the 2018/19 Ethiopia Socioeconomic Survey. 5