The Effects of Decentralized Tax Collection on Citizen Engagement

We exploit random variation in whether local bureaucrats known as avenue chiefs were responsible for property tax collection (treatment), or whether agents of the tax ministry collected taxes within avenue chiefs’ jurisdictions instead (control). We study how variation in the level of tax collection affects citizen engagement, collective action, and accountability pressures in the context of an anti-poverty program administered by local chiefs, in which citizens will have a chance to vote for audits of the chief and the provincial government. Additionally, we cross-randomize a tax prime at the individual level, which varies whether citizens are referred to as `citizens' or `taxpayers' in a community audit request form they receive. We will use the tax prime to examine whether the hypothesized effects are stronger in locations where the collective action decision is explicitly framed as an appeal to taxpayers.

2019-05-01

2019-12-15

Participation in collective action opportunity (submission of audit forms); perceptions of government capacity; demands for government accountability; engagement with government actors; expectation of receiving benefits; targeting of coethnics in taxation

This project builds on experimental variation in the level of government responsible for tax collection to evaluate affects on citizen engagement. Because the level of collection for the 2018 property tax campaign was assigned on the neighborhood level, there is random variation in whether local chiefs (whose jurisdictions are roughly coterminous with neighborhoods) or officials from the provincial tax ministry were charged with tax collection in 2018. We study how this existing variation impacts citizen engagement, perceptions of government capacity, and demands for accountability, across levels of government, and how these outcomes vary with ethnic characteristics.

We partner with the Division of Social Affairs (DIVAS) for the Province of Kasai Central, DRC, to provide citizens the opportunity to engage in collective action by allowing them to vote for community audits of an anti-poverty program administered at the neighborhood level. We will randomly select 20% of households within neighborhoods to receive audit forms, which will allow citizens to independently request audits of the provincial government body responsible for the program, DIVAS, and the local chief administering the anti-poverty program at the neighborhood level. The audits will be conducted by well-known and respected local civil society organizations: RIAC (the Network for Transparency and Anti-corruption), which specializes in promoting transparency and fighting corruption, and SOCICO (the Civil Society of Congo), which focuses on government accountability in the areas of violence, conflict, and elections. We will measure how participation in this collective action opportunity varies by the level of government responsible for tax collection, at the neighborhood level.

We also cross-randomize a tax prime at the individual level (within neighborhoods), which varies whether citizens are referred to as `citizens' or `taxpayers' in a community audit request form they receive. We will assess whether explicitly framed appeals to taxpayers influence the collective action decision.

All randomizations will be done in Stata unless stated otherwise in our pre-analysis plan.

Tax collection responsibility was randomized on the neighborhood level. Cross-randomized intervention assigned at the individual level.

Yes

161 neighborhoods

This varies by the outcome variable, as described in detail in the PAP. The estimated sample for the collective action outcome should have a sample size of approximately 4000 individual level observations. The estimated sample for survey outcomes is more than 2000 individual level observations.

See pre-analysis plan.

PAP_FINAL_SUBMITTED.pdf