On November 11, SIS professor Agustina Giraudy and her team launched the Subnational Politics Project (SPP), a new initiative dedicated to the systematic study of subnational politics in Latin America. By providing consistent, high-quality, and easily visualized data, the project helps scholars and policymakers alike compare cities or states across Latin America on a wide range of topics, including federalism, decentralization, subnational democracy and authoritarianism, party competition, electoral accountability, and territorial governance.

To learn more, we asked Giraudy about how her research informed this project, why it’s unique in the field, and what her plans are for growing the SPP.

Where did the idea for the Subnational Politics Project come from, and how does it build on your previous or current research?

I am a political scientist, and my area of expertise is Latin America—particularly, subnational politics, which studies the political, economic, and social processes that unfold beyond the national state.

My line of research has been on subnational democracy—or the lack thereof—in Latin American countries. Many of these countries transitioned to democracy from authoritarianism in the 1980s, but democracy did not trickle down evenly in subnational states and provinces. Think of the US during the Jim Crow Era. Many states continued to be ruled under authoritarianism even though at the national level, the countries were democratic. I’ve been doing research for years on different aspects of the lack of democracy at the subnational level: my first book that came out in 2015 (Democrats and Autocrats) was about that, and my second book from 2019 (Inside Countries) was more broadly about subnational politics, not just subnational authoritarianism.

One of the big challenges of doing work at the subnational level is that it’s very hard to find data, especially in Latin America and other countries in the global south. It’s especially hard to find comparable data for all subnational units within a given country. For example, in Mexico there are 32 states, in Argentina 24 provinces, and in Brazil 26 states, so getting comparable data on anything for each of these states, and over time is very complicated. For people trying to do subnational politics research, it was particularly challenging because every time a researcher wanted to study a topic, they had to reinvent the wheel, they had to collect data from scratch, in part because there was not a hub where all that data was offered.

So, one of my biggest dreams as I was working on my dissertation and then on my books and other work analyzing the subnational level was to have a website where you could go and find systematized political data that spanned a long period of time, for all states and provinces in Latin America.

What is the Subnational Politics Project, and how is it unique?

This project is a very long-term effort to collect, compile, generate, and disseminate systematic, transparent, and publicly accessible data on political institutions, elections, electoral outcomes, and more across three federal Latin American countries. It covers more than 25 years (depending on the country) with data on every single state in Mexico, Argentina, and Brazil. There’s nothing like this in comparative politics!

A major added value of this project is that you can visualize the data. You have all of the data available, such as the electoral results for all of the subnational states/provinces in these countries from the 1980s up to 2025, but you can also visualize it in map or graph form. It’s user-friendly! For instance, if you want to know the gender of the governor in each Mexican state, you have the map of Mexico and you can click “play” in the timeline and you can see the evolution: any time you see a purple state, that’s when they had a female governor. You can also see ideology of the governing party, whether elections were happening concurrently with the national elections, or even specific information about legislative electoral results. Even though this seems basic, it took years of compiling the data and building this dashboard.

Another very important thing that this tool does is that it’s the first time people can compare cities or states across the Americas. For example, in the context of trying to understand democratic erosion or backsliding, you also may want to understand what’s happening at the state level. So, let’s say you want to see the level of subnational democracy using the SUR index (subnational undemocratic regime index) for the capital cities of the three countries. You can use the graph tool to see that how a city in Argentina compares to other cities in Mexico and Brazil on its levels of democracy. And you could also pick provinces or states to levels of subnational democracy or other variables.

Before this data existed, you couldn’t do this. Measures of democracy weren’t comparable across countries because they weren’t measured in the same way for each state. This data opens up a new agenda, which is an agenda of subnational comparison across states and provinces in the Americas. If you aggregate all the states/provinces in these three countries, we have around 90 subnational units measured the exact same way. Of course, democracy indexes are only one of the data variables—using our dashboard you can compare elections, composition of legislatures, and so many other things.

Who do you hope uses this data?

Importantly, all of this data is public and free, which makes it accessible. The SPP is made up of different datasets.  If you go to databases on the dashboard and want to see the subnational democracy index, for example, you can click on the Database tab and it takes you to a repository of data housed at Harvard Dataverse where you can see and download the data. For each dataset we have, you can download it for free. According to most of my colleagues in this space, the SPP is an act of deep generosity! It’s not very common for such data to be available for anyone who wants to use it.

I hope that many people use the SPP and our dataset—not just people in academia but journalists, policymakers, or NGOs, too. If a policymaker wanted to implement a gender policy in the province of Cordoba in Argentina, they could track what their neighbors did to help them think through policy design.

We could have produced the dataset for academics only, but not everyone knows how to use a dataset, and the data visualization tool is friendlier. I’ve made my daughters use it! They provided so much feedback and we’re still working on improving the usability so that anyone can use it.

How collaborative has this project been, and what have you learned from it?

I’ve been coordinating a group of fantastic young researchers, including a postdoc and research assistants from Latin America and Germany. The SPP wouldn’t have been possible without them. It has been a real pleasure to coordinate extremely talented people who were located all over the world!

For me, it has been amazing to learn about how to make this dataset visual. I knew a lot about subnational datasets and subnational politics, but I have learned a ton about how to turn data into friendlier visualization and dashboards. It’s been a lot of work but it has been extremely rewarding! This is the work I’ve wanted to do—this idea has been with me for 10 years. During my dissertation and throughout my time at AU, I have thought about how great it would be to have a single place, a hub to go where data would be systematized and you could download it and actually start your research instead of just spending months collecting data.

This project was supported by Tecnológico de Monterrey in Mexico that is also interested in producing data like this, so I’ve been lucky to get that support.

What are your goals for the project moving forward?

This project is ever-evolving, and we want to keep expanding it: not only the usability of the dashboard in terms of expanding the coverage of the time period and variables we’re adding (we’re now in the process of collecting data for 2025), but also the “services” we could offer to people on the site using the data.

I would also love to organize a workshop on data collection. Most of the data we collected is a result of official reports that we organized and systematized, but one of the things I believe could be very interesting to discuss how to produce new subnational data with all the new technological tools we have available now, including satellite imagery, LLM to code data from reports produced by states and provinces, etc. The SPP could be a launching pad for that: how can we think about data collection or data generation for subnational units in the global south now that we have AI, satellite imagery, and other newer tools. So, I would love to organize a small workshop with colleagues from different field about how to innovate in terms of data collection in countries in the Global South.