Essays on public finance in developing countries

Overbeck, Daniel

PDF
Overbeck2025thesis.pdf - Veröffentlichte Version
Download (6MB)

URN:	urn:nbn:de:bsz:180-madoc-703797
Dokumenttyp:	Dissertation
Erscheinungsjahr:	2025
Ort der Veröffentlichung:	Mannheim
Hochschule:	University of Mannheim
Gutachter:	Janeba, Eckhard
Sprache der Veröffentlichung:	Englisch
Einrichtung:	Fakultät für Rechtswissenschaft und Volkswirtschaftslehre > Finanzwissenschaft u. Wirtschaftspolitik (Janeba 2004-)
Fachgebiet:	330 Wirtschaft
Freie Schlagwörter (Englisch):	public economics , development economics , firms , taxation
Abstract:	In most of what is commonly known as the developing world, tax collections are low. While rich OECD countries collect about a third of their GDP in taxes on average, countries in Sub-Saharan Africa or South Asia -- the most populous regions of the world -- only collect about 16% and 12\% of their GDP, respectively (World Bank, 2025; ATAF, 2018). Low tax revenues can hinder public goods provision through e.g. inadequate investments into infrastructure or funding of welfare programs. They also increase states' dependencies on external lenders and aid. The recent closure of USAID, which humanitarian organizations expect to "kill millions" (Caritas, 2025), is a timely reminder that for lower-income countries, domestic resource mobilization through taxation is imperative. Beyond generating revenues, taxation represents a key determinant of how states and societies function and also how these two interact. Tax compliance is a regular aspect of firms' and individuals' economic lives while the ability to tax and redistribute -- in other words, managing public finances -- is one of the most important function of governments. This includes incentivizing certain (firm) behaviour by selectively increasing or decreasing prices through taxes or tax exemptions and subsidies, respectively. Relative to what is known for rich countries, the economic literature has yet provided only limited evidence on the role and the effects of taxes and subsidies in lower-income countries. While this represents an emerging research area, the scarcity of evidence holds particularly true for the poorest countries. This thesis advances our understanding of public finance along three dimensions. Chapter 1 uncovers the existence of a hybrid tax system in Zambia -- one of the least developed countries in the world (United Nations, 2023) -- where instead of following the legal tax schedule, tax payments are simply determined through bargaining with tax collectors. Chapter 2 shows how the implementation of Special Economic Zones in India, which are essentially place-based tax exemptions for firms, led to sectoral shifts from agriculture to manufacturing and service industries in nearby villages. Chapter 3 analyzes how firms respond to the South African carbon tax, which is the first of its kind in any developing country. Each chapter is self-contained. In Chapter 1, which is joint work with Eliya Lungu, I study the taxation of small firms in Zambia. Their taxation is highly relevant as small firms make up 80-90\% of taxpayers and account for about $40\%$ of GDP in the average developing country. I apply empirical as well as theoretical methods to show that bargaining over tax payments is an important feature of tax compliance and enforcement in lower-income countries such as Zambia. The empirical analysis is twofold. First, analyzing the universe of administrative tax filings from Zambia, I document sharp bunching in (i) dominated regions above tax schedule discontinuities, inconsistent with standard models of tax compliance and (ii) at round number tax payments, implying that certain payments are being targeted. Second, I provide additional qualitative evidence from a survey I conducted of Zambian firms via an external provider. The findings suggest that discussing tax payments with tax officials before filing taxes is widespread, in line with tax payments being the outcomes of bargaining. Such bargaining over taxes is consistent with fact (ii), as bargaining outcomes are often round and salient numbers, and with fact (i), because tax schedule discontinuities restrict the set of feasible bargaining outcomes. The theoretical model I propose generalizes the conventional Allingham & Sandmo (1976) model to allow for bargaining as a mode of tax compliance. I show that as long as state capacity is low, which is the case for developing countries, bargaining over taxes leads to Pareto-improvements for both taxpayers and the state. This result rationalizes why bargaining is observed for small firms in low-income countries but not for larger firms in more advanced economies. Chapter 2 is joint work with Johannes Gallé, Nadine Riedel and Tobias Seidel and has been published in the Journal of Public Economics (c.f. Gallé et al., 2024). In this chapter, I study the impact of the implementation of Special Economic Zones (SEZs) in India. According to the UNCTAD (2019), the total number of SEZs worldwide increased from 500 in 1995 to about 5,400 in 2018 - with the vast majority of the new zones being located in developing economies. Therefore, they represent a highly prevalent policy tool. I quantify the local economic impact of SEZs established in India between 2005 and 2013 econometrically using a spatial difference-in-differences design. Relying on a novel dataset that combines census information on the universe of Indian firms with geo-referenced data on SEZs, I find that the establishment of SEZs increased local manufacturing and service employment, with positive spillovers up to 10 km from the SEZ area. The analysis shows that the gains in manufacturing and service employment were accompanied by a decline in agricultural labor, especially for women, suggesting that the policy contributed to structural change. In further analysis, I document that significant local employment effects occur across different types of SEZs: privately and publicly run zones, and SEZs with different industry designations. A back-of-the-envelope calculation shows that the costs of the program (i.e. foregone tax revenues) slightly exceed the welfare gains from additional employment. Overall, I still interpret the findings to dispel the general pessimism about zone programs in developing countries outside of China. Chapter 3, which is coauthored with Johannes Gallé, Rodrigo Oliveira, Nadine Riedel and Edson Severnini, concerns the role of taxation for environmental policy. It provides the first comprehensive analysis of how firms in emerging economies respond to carbon taxation, leveraging detailed administrative data from South Africa -- a potential trailblazer for other developing countries with limited state capacity amid the growing global push for carbon pricing. I examine the dynamic impacts of the carbon tax on firm-level outcomes -- such as profits, sales, capital, and labor inputs -- across manufacturing and mining firms, which are key sectors in the context of the carbon tax. Contrary to concerns that carbon taxes may hinder economic growth or reduce employment, the findings show no evidence of negative average impacts on firm performance or jobs. However, this overall result masks significant heterogeneity in the tax's effects across sectors, driven by the sector-specific design elements of the South African carbon tax. Firms expecting higher effective tax rates may have intensified their use of emission-intensive machinery and depreciated capital in anticipation of the tax. This behavior appears to stem from firms resolving regulatory uncertainty or seeking to recover costs from stranded assets. In conclusion, this thesis make three main contributions to the literature on public finance in developing countries. First, it shows how tax compliance of small firms deviates strongly from the predictions of standard economic models. Second, it highlights how place-based tax exemptions for firms can lead to beneficial outcomes for surrounding regions. Third, it provides evidence that carbon taxation, which is high up on policy agendas, may not be as harmful to economic performance as expected. These novel findings might inform policymakers on a broad variety of public finance issues and importantly, show how optimal policy recommendations may need to be adjusted to account for the institutional realities of developing countries.