Tax Dispersion and the Spatial Allocation of Economic Activity

Edward Davenport and Tishara Garg

This project has been retired

Small Research Grant (PhD)

Firms choose where to locate across space for a variety of reasons. For example, sector pairs with strong input-output linkages (e.g. shoes and leather) tend to co-locate in many countries to reduce transport costs. However, if firms within a given sector face different tax rates across space, even within a country, this may distort their location choices and have consequences for aggregate efficiency and welfare. On the other hand, central governments in many lower-income countries may not necessarily want to harmonise taxes across space, perhaps because of a desire to redistribute across regions or because of political economy constraints. This project studies this issue both theoretically and empirically. Theoretically, the main research question asks under what conditions is it optimal for a central government to harmonise taxes across space. Empirically, the project’s focus is on India, where, prior to a large reform in 2017, each state had discretion in setting sector-specific tax rates. The primary empirical research question asks whether and to what extent the 2017 harmonisation brought the Indian economy closer to the theoretical optimum.

First, for the theoretical section, the research team develops a Ramsey model of commodity taxation that incorporates: (i) a potential preference for the central government to redistribute across regions; and (ii) potential constraints on the central government on their ability to change existing local taxation policies. Second, for the empirical section, they use a quantitative model of trade (that will be a special case of the more general, Ramsey model). This is important as, in the empirical setting: (i) taxes change across all locations at the same time; and (ii) it makes it possible to study aggregate welfare impacts. The team uses this quantitative model to study counterfactuals, one of which will be the 2017 harmonisation while another will be guided by the theoretical findings.

The research will provide insight into the trade-offs involving in having harmonised versus varying taxes across regional units within a country. On the one hand, harmonising taxes may improve aggregate efficiency, for the reasons discussed above. If the central government is able – and willing – to redistribute the excess revenue, this is likely Pareto improving. On the other hand, if the government cannot do this, or is unwilling to, then it may be optimal to have non-uniform taxes. While this project’s empirical focus is India, which has a somewhat unique system tax system, the primary question of tax decentralisation is general. For example, Fjelstad, Chambas, and Brun (2014), in a review of local government taxation in sub-Saharan Africa, argue that these are likely to generate economic distortions, tax competition, and poor intra-governmental coordination.

Research Team

Related content

STEG Working Paper

Bureaucratic Nepotism

Juan Felipe Riaño • Research Theme 5: Political Economy and Public Investment
STEG Working Paper

Occupational Inheritance in Africa

Nicolas Syrichas • Research Theme 2: Labour, Home Production, and Structural Transformation at the Level of the Household