Marketing Taxation: Experimental Evidence on Tax Compliance in Malawi

• In Event: Field Experiments on Service Delivery and Accountability

Abstract

Taxation has historically played a critical role in state development. Without taxation, many states struggle to secure their borders, provide public goods, and invest in economic development; taxation has also been linked to democratization, and to good governance in general. This makes increasing tax compliance a critical task for developing countries. However, there is limited experimental evidence on how best to improve tax compliance in these settings, especially whether it is more effective to improve enforcement and tax collector incentives, or to seek quasi-voluntary compliance through bargaining with taxpayers.

This paper presents the results of a tax compliance field experiment conducted in 128 markets across 8 districts in Malawi. In Malawi, as in many sub-Saharan African countries, fees from open-air markets are one of the largest sources of “own” revenue for local governments. However, current tax compliance levels are low, limiting the ability of local governments to fund service provision or invest in higher bureaucratic capacity to collect more taxes. Potential taxpayers, in turn, are reluctant to pay taxes until they see their ‘return on investment’ in the form of government-provided public services. To address this cycle of low taxation, limited service provision, and low government capacity, we use a 2x2 factorial design to compare the effectiveness of two intervention bundles. The “top-down” bundle aimed to increase compliance and revenues by setting revenue targets for tax collectors; providing rewards for markets that met their targets; and introducing a system by which collected taxes could be deposited via mobile money, rather than physically taken to the district. The second, “bottom-up” bundle was designed to improve quasi-voluntary compliance through facilitating communication and tax bargaining between vendors and government, improving market services through small construction projects, and introducing an SMS communication system to increase the transparency surrounding tax revenues.

We find that markets receiving the bottom-up bundle of interventions have significantly higher tax compliance at endline, compared to the control group, as measured by whether vendors can produce a recent tax receipt. We also find higher endline levels for self-reported tax compliance and revenues in the bottom-up treatment markets, although these results may be due to baseline imbalance and missing data. Mechanism tests show that the intervention improved tax morale and vendors’ satisfaction with services. In the top-down treatment group, we find few effects on overall tax compliance and revenues, but some evidence of higher tax collector effort, including enforcing consequences for tax defaulters. Finally, negative treatment effects are seen in the markets that received both the top-down and bottom-up interventions, suggesting that there may be interactions between the top-down and bottom-up bundles.

Together, these results suggest that enhancing accountability and transparency surrounding tax revenues may increase citizens’ willingness to pay taxes. This provides a potential solution to the problem of how to increase revenue collection in low capacity environments.

Authors

• Lucy E. S. Martin, University of North Carolina, Chapel Hill

• Brigitte Seim, University of North Carolina-Chapel Hill

• Simon Hoellerbauer, University of North Carolina, Chapel Hill

• Luis A. Camacho, NORC at the University of Chicago