close× Call Us
Title Exploring Social and Group Dynamics of Tax Compliance: A Conjoint Experiment Among Zambian Taxpayers
Post date 07/31/2018
C1 Background and Explanation of Rationale

An efficient and well-functioning tax system is a necessary foundation for any country to achieve economic development. One of the key features of a good tax system is the level of tax compliance amongst economic players. In Zambia, tax compliance rates are very low, especially among the Micro, Small and Medium Enterprises (MSMEs). Filing compliance among these taxpayers has been declining since 2010. In 2016, taxes paid by MSMEs only contributed about 15% of the Zambia Revenue Authority’s (ZRA) collections. As part of a multi-phase research project, we conduct a survey that unpacks the barriers to tax compliance among this group of taxpayers. Along with exploring many of the oft-cited barriers to tax compliance, such as low rates of monitoring and enforcement of the tax law, a conjoint experiment sheds light on the social and group-level dynamics of tax compliance and a list experiment refines our measures of tax compliance. Here, we present the pre-analysis plan for the conjoint and list experiments.

C2 What are the hypotheses to be tested?

Hypothesis 1: MSME operators who rely more on public services will be more likely to pay taxes than those that who rely less on public services.
Hypothesis 2: MSME operators are more willing to pay their taxes when norms supporting tax-paying are high.
Hypothesis 3: MSME operators are more willing to pay their taxes when business associations act as tax collectors.
Hypothesis 4: MSME operators are more willing to pay their taxes when other business owners will learn about their tax compliance.
Hypothesis 5: MSME operators are more willing to pay their taxes when audits are more likely.

C3 How will these hypotheses be tested? *

To test Hypothesis 1, we will examine the difference in means across the list experiment treatment groups, by reliance on public services. We will follow Hainmueller et al. (2015) in conducting the conjoint experiment analysis to non-parametrically estimate the effects of each conjoint dimension, simultaneously testing hypotheses 2-5.

C4 Country Zambia
C5 Scale (# of Units) Approximately 1250
C6 Was a power analysis conducted prior to data collection? No
C7 Has this research received Insitutional Review Board (IRB) or ethics committee approval? Yes
C8 IRB Number IRB00011000
C9 Date of IRB Approval 04/04/18
C10 Will the intervention be implemented by the researcher or a third party? Researchers
C11 Did any of the research team receive remuneration from the implementing agency for taking part in this research? not provided by authors
C12 If relevant, is there an advance agreement with the implementation group that all results can be published? Yes
C13 JEL Classification(s) H25, H26