police mexi BBC

Nina Robinson/BBC

Researcher     Gustavo Flores-Macías
Location Mexico
Duration September 2013 - January 2014


In spite of the importance of taxation for the provision of public safety, the relationship between the two remains understudied. Although recent studies have emphasised the importance of taxation for development (Inter-American Development Bank 2013) and building state institutions more generally (Brautigam et al. 2008), we know little about how security conditions are related to taxation.

Since governments are more likely to encounter opposition to greater tax burdens than greater oversight or decreasing rates, a key goal of this study is to understand how to ameliorate such opposition and how different factors might change people’s views. 

This approach is novel, not only in its examination of the security-taxation nexus, but also in that it places emphasis on studying the public’s attitudes toward increased taxation instead of the conventional focus on compliance (e.g. Cummings et al. 2009). To do so, the study relies on experimental survey methods based on a nationally representative sample, which improves on the external validity of the majority of tax-related experiments administered to small convenience samples. 

The need for understanding the key factors behind attitudes toward taxation in developing countries (Fjeldstad et al. 2012) serves as the motivation for this study. In particular, the study’s objective is to evaluate the relationship between public safety and taxation.

This research relies on a combination of experimental survey design and public opinion research. Experiments allow researchers to vary certain treatment conditions that may not be possible to observe simultaneously in a particular context, for example, one cannot observe individuals’ willingness to pay taxes with and without certain rules at the same time.

Additionally, when properly randomized, they allow researchers to attribute causal effects to experimental manipulations and determine the magnitude of these effects (Druckman et al. 2006). This design allows us to transcend correlations-based research that has characterised public opinion studies, while responding to the need for research on the key determinants of tax attitudes in developing countries (Fjeldstad et al. 2012).


ICTD Working Paper 51 - Building Support for Taxation in Developing Countries: Experimental Evidence from Mexico by Gustavo A. Flores-Macías.

The dataset is also now available here.