Research

With: Pia Rattenhuber

Journal: International Journal of Microsimulation, 2019

Abstract: We use four novel, cross-country comparable tax-benefit microsimulation models for Ecuador, Ghana, Tanzania and South Africa to analyse ex ante the expansion of a universal old-age pension in a static setting. Universal pensions would significantly reduce poverty and inequality in settings where no means-tested old-age pensions exist (such as in Ghana and Tanzania). If means-tested old-age pensions exist and shall be maintained, universal pensions as a top up scheme only make a difference for the income distribution if the existing schemes do not reach the entire vulnerable population such as in Ecuador. Costs for the proposed schemes are substantial. 

With: Tina Kaidu Barugahara, Joseph Okello Ayo, Jukka Pirttilä, and Pia Rattenhuber

Journal: International Tax and Public Finance, 2025, Vol. 32, p. 1177-1212 

Abstract: The extent of redistribution in low-income developing countries, including in Africa, is very limited, which raises the question whether the tax rates of high-income individuals should be raised. A crucial parameter when considering a potential increase in progressivity is the response of taxable income to increased tax rates. In this paper, we evaluate a major personal income tax reform in Uganda that came into effect in 2012–2013, which increased the top tax rate by 10 percentage points. Using the universe of pay-as-you-earn (PAYE) administrative data from the Uganda Tax Authority, we analyse the impact of the reform on reported labour incomes. In the preferred specification, we find very limited support for behavioural reactions. However, heterogeneity analysis reveals that top-income workers in firms handled by ordinary (as opposed to medium or large taxpayer) offices report lower incomes after the reform. We also find suggestive evidence that part of the response may arise from income shifting to other tax bases. The reform managed to raise more revenue and it also led to a limited reduction in after-tax income inequality.

Editors: Maria Jouste, Ravi Kanbur, Jukka Pirttilä and Pia Rattenhuber

ESTIMATED FOR OPEN ACCESS ONLINE PUBLICATION IN NOVEMBER 2025 | This book examines the role of social protection and taxation systems in developing countries during times of crises. The main objective of the work is to promote understanding about proper crisis response, and the way social protection and tax systems can be made more sustainable to support countries' paths through and out of economic crises. 

While there is a vast body of literature evaluating social protection programmes in general, few of these focus on episodes of crisis. During the COVID-19 pandemic, the number and scale of social protection programmes increased rapidly around the globe, but many developing countries lacked a systematic approach to social protection and taxation.

Crises have a devastating effect on employment, incomes, and livelihoods. These impacts are more dangerous when people are already in a vulnerable position. This has become very clear during the latest worldwide crises, including the sharp rise in food and fuel prices. 

Armoured with comprehensive policies and systems for social protection and taxation, developing countries could enable automatic and speedy assistance when crises hit across the entire income distribution. Analogously to the currently standard financial stress testing in anticipation of crises, thorough stress testing of social protection and tax policies is also of crucial importance.

Poor Protection explores to what extent tax-benefit systems can act as automatic stabilizers in a developing country context during crises, and how they can help provide the fiscal space necessary to cushion at least the most detrimental developments in terms of inequality.

Chapters:

Chapter 1: Introduction: protecting the poor and vulnerable when crises are the New Normal

Authors: Jukka Pirttilä, Maria Jouste, Ravi Kanbur and Pia Rattenhuber

Chapter 10: The potential of universal basic income schemes to buffer shocks: comparing Uganda and Zambia during COVID-19  

Authors: Enrico Nichelatti, Maria Jouste and Pia Rattenhuber

Chapter 13: Summary

Authors: Jukka Pirttilä, Maria Jouste, Ravi Kanbur and Pia Rattenhuber

Chapter 14: Policy takeaways

Authors: Jukka Pirttilä, Maria Jouste, Ravi Kanbur and Pia Rattenhuber

With: Enrico Nichelatti and Pia Rattenhuber

Abstract: The debate over universal basic income (UBI) has gained traction in the developing world in recent years. We analyse the effects of four UBI schemes on poverty and inequality measures during normal times and times of crisis in Uganda and Zambia. We use static microsimulation models and nationally representative household surveys for each country. Our results show that in Zambia, where the existing social protection benefits have more extensive coverage, the least generous UBI benefit leads to higher poverty and inequality compared to existing benefits. By contrast, in Uganda, a country with only one notable social protection programme, all UBI scenarios reduce poverty and inequality. Yet, differences between welfare estimates for the existing systems and UBI scenarios whether in normal or crisis times are not large when UBI policies are revenue neutral or expenditures calibrated to the regional average. In normal times and in times of crisis, poverty reduction increases with the generosity of the UBI benefit amount in both countries. UBI schemes clearly outperform existing systems only with UBI benefit amounts that result in unrealistic expenditure levels.

With: Milly Isingoma Nalukwago and Ronald Waiswa

Abstract: This paper conducts an impact evaluation of the effects of two tax administration interventions—a taxpayer register expansion and education programme, and a new electronic filing system for presumptive tax—on the number of small business taxpayers and presumptive tax revenues in Uganda. Using a difference-in-differences approach and administrative data covering both presumptive taxpayers and comparable small corporate income taxpayers, we find that the number of small business taxpayers filing tax returns and presumptive tax revenues increased substantially after the interventions. We argue that the interventions complement each other because both interventions were established around the same years, and the taxpayer register expansion programme focused on not only registering but also educating taxpayers with regard to tax compliance. We analyse the cost-effectiveness of the taxpayer register expansion programme and find that the benefits outweigh the costs.

With: Jesse Lastunen, Pia Rattenhuber, Kwabena Adu-Ababio, Katrin Gasior, H. Xavier Jara, David McLennan, Enrico Nichelatti, Rodrigo C. Oliveira, Jukka Pirttilä, Matteo Richiardi, and Gemma Wright

Abstract: This paper analyses the distributional effects of the COVID-19 pandemic and related tax-benefit measures in 2020 in a cross-country comparative perspective for five African countries: Ghana, Mozambique, Tanzania, Uganda, and Zambia. We first estimate the impact of the crisis on disposable incomes, how effects vary across the income distribution, and in how far tax-benefit policies stabilized earnings losses. We then evaluate the impact on income-based poverty and inequality and the contribution of discretionary tax-benefit policies in alleviating the shock. Our analysis shows modest increases in headcount poverty rates and inequality, and somewhat larger effects on the poverty gap due to lower relative earnings losses of the poor population at the early stage of the pandemic analysed here. We find very limited stabilizing power of tax-benefit policies overall and automatic stabilizers in particular. This illustrates gaps in coverage for the large informal sector and a general lack of income-related means-tested benefits. Except for the Emergency Social Cash Transfer in Zambia, discretionary tax-benefit policies adopted in response to COVID-19 have had limited impact. Pausing a large school feeding programme in Ghana during lockdown has in turn put additional pressure on households with school-age children. 

With: Katrin Gasior, David McLennan, Mbewe Kalikeka, and Miselo Bwalya

Abstract: A large share of the population in Zambia is living below the national poverty line. To reduce poverty, in 2019, the government initiated the Cash Plus reform, which aims to build on the existing Social Cash Transfer as a floor benefit with additional benefits to take account of the multidimensionality of poverty. We use the tax–benefit microsimulation model MicroZAMOD to analyse the coverage and poverty impact of the current social protection system and to assess the extent to which potential Cash Plus reform scenarios can improve the status quo. The results highlight the need for reform to achieve greater poverty reduction. Overall, coverage of the extremely poor is high but coverage by the Social Cash Transfer as the envisaged floor benefit in the Cash Plus reform is low, and the benefit amount is often too little as it does not take account of household composition. In theory, the Cash Plus reform offers the potential to achieve a greater poverty impact through multiple support. However, the simulations of the potential reform scenarios show that this requires more than the proposed Cash Plus design.

With: Sebastian Ssebuyira

Abstract: This technical note describes the Uganda Revenue Authority registration data covering all individual and non-individual registered taxpayers between 2009 and 2024. The data contains 33 variables and is divided into two datasets: 1) taxpayer register, and 2) tax type register. These datasets include taxpayer characteristics and information on registration for different tax types. This data is mergeable with tax return data, and alone, it will not contain taxpayer’s income information. The data is structured to follow the information collected from individuals and non-individuals submitting the Taxpayer Identification Number Registration form. This note describes the main elements of the data and provides a list of variables. Finally, the note provides some summary statistics of the data.Technical Not

With: Gerald Agaba and Sebastian Ssebuyira

Abstract: This technical note describes the Uganda Revenue Authority (URA) corporate income tax panel constructed from corporate income tax (CIT) returns and firm registration data for the financial years 2013/14 to 2022/23. The panel dataset contains over 300 variables, which allow the user to study, for example, firms’ balance sheets, debt structure, capital investments, and tax liabilities. This represents the richest data source available for studying the behaviour of Ugandan firms. It is structured so as to follow the order of the URA’s Non-Individual Income Tax Return form, and this note allows the user to trace variables to their location in the form. This note describes the variable content of the data and the process of constructing the dataset and documents how various challenges were tackled, including the cleaning process. Finally, the note provides some summary statistics of the data.

With: Quynh Tieu and Joseph Okello Ayo

Abstract: This technical note describes Ugandan trade data covering import and export declarations for ten calendar years from January 2013 to December 2022, constructed from transactional-level Ugandan administrative data. The trade data series contains transaction-level ten datasets on imports and ten on exports declarations from January 2013 to December 2022 under different import and export regimes. The trade data allowthe user to study various concepts at the transaction-level: the imports and exports volume, the characteristics of imported and exported/re-exported goods, the customs value/price amount, the amount of tax/fee/surcharge/license assessed and applicable to the declared goods. In terms of structure, this technical note first describes the variable content of the institutional settings of the customs system in Uganda, then the coverage for the 20 datasets, and the process of constructing the datasets and their content. Lastly, we also present some summary statistics of the datasets, the timeline to update the data, and some potential research avenues using the data. 

With: Gerald Agaba and Sebastian Ssebuyira

Abstract: This technical note describes the Uganda Revenue Authority (URA) presumptive tax returns data covering financial years from 2015/16 to 2022/23. The data contains 21 variables, including information on taxpayers’ characteristics, turnover, and payable taxes. It covers the population of small businesses eligible for a presumptive tax regime that file their incomes using the URA Income Tax for Small Businesses form. This data can be merged with other available administrative tax data, such as the URA firm panel, and thus provides a broader picture of businesses’ performance and their behaviour in Uganda. This note describes the key elements and the construction of the data and presents the variable description. Finally, the note provides descriptive statistics of the data.

With: Quynh Tieu, Gerald Agaba, and Brian Arinaitwe

Abstract: This technical note describes the Uganda Revenue Authority (URA) Pay-As-You-Earn (PAYE) data covering financial years 2013/14 to 2021/22. PAYE is paid by an employer who withholds personal income tax on employees’ wages and other employment incomes like benefits and allowances. The data are constructed from monthly PAYE tax returns and assessments and contain 30 variables, including information on employees’ incomes and taxes and employers’ characteristics. These data can be merged with other available administrative tax data, such as the URA firm panel. Thus, it can be used to study formal employment and income distribution in Uganda. This note describes the main elements of the data and provides a list of variables. Finally, the note provides some summary statistics of the data. 

With: Kyle McNabb, Dorothy Nakyambadde, and Susan Kavuma

Abstract: This technical note describes the Uganda Revenue Authority (URA) firm panel, which is constructed from administrative corporate income tax (CIT) returns and firm registration data for the financial years 2013/14–2019/20. The panel dataset contains over 300 variables, which allow the user to study, for example, firms’ balance sheets, debt structure, capital investments, and tax liabilities. This represents the richest data source available for studying the behaviour of Ugandan firms. It is structured so as to follow the order of the URA’s Non-Individual Income Tax Return form, and this note allows the user to trace variables to their location in the form. This note describes the variable content of the data, the process of constructing the dataset and documents how various challenges were tackled, including the anonymization and cleaning process. We present some summary statistics of firms captured within the dataset.

With: Mbewe Kalikeka, Miselo Bwalya, Katrin Gasior, and David McLennan

Abstract: This report documents MicroZAMOD, the SOUTHMOD model developed for Zambia. This work was carried out by Zambia Institute for Policy Analysis & Research (ZIPAR) in collaboration with the project partners. The results presented in this report are derived using MicroZAMOD version 2.12 running on EUROMOD software 3.1.8. The report describes the different tax–benefit policies in place, how the microsimulation model picks up these different provisions, and the database on which the model runs. It concludes with a validation of MicroZAMOD results against external data sources. For further information on access to MicroZAMOD, see the model page. The MicroZAMOD model and its documentation in this country report has been prepared within the UNU-WIDER project on ‘SOUTHMOD—simulating tax and benefit policies for development’. For more information, see the SOUTHMOD project page.

With: Katrin Gasior, Helen Barnes, Jesse Lastunen, David McLennan, Michael Noble, Rodrigo Oliveira, Pia Rattenhuber, and Gemma Wright

Abstract: This technical note presents a modelling approach used in Lastunen et al. (2021) where tax and benefit policies are scaled to reflect their actual duration during a single calendar year. It can be applied to tax-benefit microsimulation models implemented in the EUROMOD software. The method is particularly useful when evaluating the impact of discretionary policy measures adopted during the COVID-19 pandemic. Tax-benefit microsimulation models typically simulate policies at a specific point in time, which is problematic when considering changes in policies over time, for instance over the course of the coronavirus pandemic. Using the standard point-in-time approach, only those policies that were in place at the specific cut-off date (e.g. July 2020) would be considered and, even if short-lived in reality, assumed to be effective throughout the whole calendar year. The approach presented here applies ‘full-year adjustment’ to any policies implemented in 2020 that were in force for less than 12 months, ensuring that relevant benefit amounts and tax liabilities are scaled to reflect realistic payments during the year.