Comparing tax and benefit microsimulation models: a guide for analysts in low- and middle-income countries

Distributional analysis is a crucial component of policymaking. Distributional models estimate who gains from and who bears the costs of different tax and benefit policies. They can provide a detailed diagnostic assessment of the existing fiscal system, help identify areas for reform, and model the impacts of proposed reforms and alternative policy scenarios. This helps policymakers understand the distributional, poverty, and inequality impacts of different policies and their combinations.

Multiple tools for distributional analysis are now available for low- and middle-income countries (L&MICs). This report by TaxDev researchers compares three selected initiatives that have developed comprehensive tools covering multiple tax and transfer policies in several L&MICs: SOUTHMOD and TaxDev microsimulation models, developed by researchers at UNU-WIDER and the IFS, respectively; and CEQ fiscal incidence models developed by the CEQ Institute and/or the World Bank. It aims to support researchers, analysts, and policymakers to better understand commonalities and differences between these models.

Key findings:

• The models differ in their focus and analytical approach. CEQ-style fiscal incidence analysis is well-suited for a detailed assessment of the existing system using different methods of allocating existing taxes and transfers. Microsimulation models by SOUTHMOD and TaxDev are ideal for simulating counterfactual policy proposals.
• The initiatives target different users. SOUTHMOD and CEQ focus on collaboration with local researchers and emphasise wider accessibility and cross-country comparability of their tools. TaxDev and the World Bank often partner with ministries of finance and revenue authorities and tailor their approach to the local needs.
• The models have a different country coverage. CEQ is available for the most countries, while TaxDev with its bespoke approach covers the smallest set of countries.
• Gradually embedding these tools in policymaking is important. It is better to start with simpler analysis to develop capacity before expanding to more complex models. In contrast, maintaining multiple overlapping models in settings with limited capacity can create confusion and harm take-up.

Find below the executive summary of the report and download the full report on the right side of this page. 

Executive summary

Tax and benefit microsimulation models (MSMs) and fiscal incidence analyses (FIAs) are important tools for understanding how fiscal policies affect households across the income distribution. They are used to assess who bears the burden of taxation, who benefits from public spending, and how alternative fiscal reforms may affect poverty, inequality, and the public finances. Interest in these approaches has grown substantially in low-and middle-income countries (L&MICs) over the past two decades, driven by improvements in household survey data, expanding administrative data systems, and an increasing policy focus around poverty reduction, inequality, and domestic revenue mobilisation.

This report compares three families of models that have been developed and applied in L&MICs: SOUTHMOD, led by UNU-WIDER and based on the EUROMOD platform; TaxDev models, developed by the Institute for Fiscal Studies (IFS); and CEQ models developed by the Commitment to Equity Institute, alongside related tools implemented by the World Bank. Although all these approaches seek to analyse the distributional effects of fiscal policy, they differ in their objectives, analytical architecture, methodological assumptions, country coverage, institutional arrangements, and intended users.

Key findings from the comparison of the three families of models

1. Country coverage differs substantially across models. SOUTHMOD currently covers 12 core countries using the EUROMOD platform, with additional country models developed by partner institutions using the same architecture. TaxDev has developed full MSMs for Ghana and Ethiopia and selected Latin American countries, alongside partial models and analytical tools for several other African countries. CEQ analyses have been implemented in more than 80 countries, while the World Bank has developed tailored MSMs for more than 50 developing countries since 2016.

    

2. MSMs and FIA differ in their main focus and analytical approach. MSMs such as SOUTHMOD and TaxDev focus primarily on simulating counterfactual reforms and supporting ongoing policy design, monitoring, and evaluation. They simulate tax liabilities and benefit entitlements by applying statutory rules to household survey data alongside incidence and compliance assumptions. FIA approaches such as CEQ focus particularly on analysing the distributive effects of existing fiscal systems and benchmarking fiscal redistribution across countries. They place greater emphasis on the allocation of existing taxes and expenditures across households using survey data, administrative aggregates, and simulation methods combined with incidence and compliance assumptions.

    

3. The models differ in the scope of fiscal instruments covered. SOUTHMOD and TaxDev MSMs generally focus on direct and indirect taxes and cash transfers and are particularly well suited for simulating policy reforms. CEQ studies provide the broadest coverage of fiscal instruments, including direct and indirect taxes, subsidies, and cash and in-kind transfers such as education and health spending. World Bank MSMs tend to focus on a narrow set of tax instruments.

    

4. All approaches rely primarily on nationally representative household surveys and use static (arithmetic) simulations. In their standard versions, they generally abstract from behavioural responses to policy changes and wider macroeconomic effects.

    

5. Methodological choices vary across models. Important differences include the treatment of social insurance contributions, assumptions about tax compliance and informality, approaches to calibration and validation, the choice of welfare measures (income versus consumption), and the way the analysis is presented.

    

6. Institutional models and intended users also differ. SOUTHMOD and CEQ place strong emphasis on collaboration with local researchers and wider accessibility outside government. TaxDev and World Bank models are often developed in close partnership with ministries of finance (MoF) and tax authorities to support specific tax reform processes, strengthen the analytical capacity of tax policy units, and address other technical assistance needs.

    

7. FIA and MSM approaches should often be seen as complementary. CEQ FIA provides the most comprehensive diagnostic assessment of the existing fiscal systems, which helps identify potential areas for reform. However, MSMs are then needed to analyse any alternative policy proposals.

    

8. Gradually embedding these tools within policymaking institutions is critical. Their long-term value depends not only on technical quality but also on whether governments develop the capacity, incentives, and institutional ownership needed to use them regularly in fiscal policymaking. Starting with simpler analysis focusing on a subset of fiscal instruments and gradually expanding the models over time is a good strategy in countries where these tools are not yet available. In contrast, maintaining multiple overlapping models in settings with limited capacity can create confusion by unexplained model differences and hence harm take-up.

The choice of model ultimately depends on the user’s objectives. For a detailed diagnosis of the existing fiscal system, analysing a broader set of fiscal instruments, and benchmarking against other countries, CEQ-style FIA may be more appropriate. For a more forward-looking agenda focused on evaluating alternative reforms to taxes and cash transfer, analysts should aim for an MSM instead. SOUTHMOD models are based on standardised methodology that enables cross-country comparisons and have a wider community of users that fosters sustainability of their tools. For a more country-specific analysis that allows for more complex modelling, a bespoke approach, as taken by TaxDev, may be preferred.

The report also emphasises that the long-term value of these tools depends not only on technical sophistication but also on whether they become embedded within national policymaking processes. Building sustainable analytical capacity within MoF, revenue authorities (RA), statistical agencies, universities, and research institutes is therefore critical. Countries do not necessarily need to begin with highly sophisticated models covering the entire fiscal system. Starting with simpler analyses focused on key taxes or transfers can already provide valuable policy insights and help build institutional familiarity with distributional analysis. Over time, models can be expanded to incorporate additional fiscal instruments, administrative data, behavioural assumptions, and indirect effects through production chains.

Ultimately, effective use of FIA and MSM tools can help governments move beyond analysing individual taxes or expenditures in isolation and instead evaluate packages of reforms, fiscal trade-offs, and the combined effects of revenue and spending measures on different population groups. As governments face increasing pressure to reconcile fiscal sustainability with poverty reduction and inequality objectives, these tools are likely to become increasingly central to evidence-based fiscal policymaking in L&MICs.