Why is the research valuable?

Municipal governments in Mexico are vital for service delivery and the development of political parties. However, municipal-level elections have been underexplored due to a lack of data. Most existing electoral datasets either cease at the state level or provide inconsistent municipal-level data across different times and locations. We have created this database to address this gap and to assist researchers, policymakers, and civic technologists in gaining a better understanding of Mexican politics.

Our research focuses on the increasing interest in subnational political analysis and causal inference at the local level. The dataset we have compiled includes electoral results from over 2,400 municipalities between 1994 and 2019, covering 31 states. We have standardized the vote counts for mayoral elections across precincts, enabling new analyses of party and coalition competition, incumbency effects, and voter turnout. This work contributes to establishing stronger empirical foundations in the study of electoral behavior and democratic accountability in Mexico.

What did the authors do?

This project emerged from both a necessity and a sense of curiosity. While researching local politics in Mexico, we observed that municipal electoral data, particularly at the precinct level, was fragmented and difficult to access. This unavailability of data complicates the statistical identification strategies used in political science, which are essential for conducting causal analyses.

To address this issue, we systematically collected and harmonized precinct-level results from state electoral authorities. In many instances, manual validation was necessary to ensure consistency across different times and jurisdictions. Our focus extended beyond just vote counts; we also examined voter turnout, the number of registered voters, the composition of party coalitions, and the affiliations of both incumbent and runner-up candidates, who often compete within these coalitions.

One of the more intensive aspects of this process was identifying the leading party within municipal coalitions, specifically to determine the partisanship of the running mayoral candidate. Since this information was not always clearly stated in official sources, we conducted an extensive internet search. This search involved reviewing local news articles, party websites, electoral campaign documents, and any public statements or materials that could indicate the identity of the party.

The final dataset includes:

• Precinct-level votes for each party and coalition in each municipal election.
• Total votes, number of registered voters, and turnout. 
• Partisan information of incumbent mayors and state governors.
  
  

We also expanded the dataset to allow for the most common identification strategies, including difference-in-differences and regression discontinuity designs. We added municipal- and state-level electoral data.

What are the implications of this study?

We believe that this dataset serves as a crucial foundation for enhancing the empirical study of local politics in Mexico. By offering detailed data across both space and time, it addresses a significant limitation in subnational research. Furthermore, it provides a platform for comparative studies in political science, development economics, and public policy.

This database supports investigations into decentralization, public goods provision, electoral reforms, gender quotas, criminal violence, and local spending—all areas where scholars have long requested better micro-level electoral data.

Looking ahead, the dataset encourages updates and extensions. Adding candidate-level information, gender breakdowns, or linking it to administrative data on budgets and outcomes could enhance its usefulness. For civic tech communities and election monitors, this dataset provides new tools to visualize patterns of political participation and assess democratic inclusiveness across Mexico.

Read the full paper

You can read the article and access the dataset following this link.