Ignacio Flores, who is currently the data architecture expert at The GC Wealth Project and previously was a Stone Center postdoctoral scholar, recently launched Distribuciones.info, a Spanish-language site aimed at providing research by Flores and his coauthors on inequality in Latin America to a general audience. Flores’s research partners in this venture include Pablo Arriagada of Our World in Data, Mauricio De Rosa of Uruguay’s University of the Republic, and Marc Morgan of the University of Geneva; all of the site’s visualizations were created by Arriagada.

Unlike many previous studies, Flores and his coauthors argue that inequality in Latin America did not decline for about a decade in the early 2000s. Flores recently spoke to the Stone Center about his goals for Distribuciones and his research findings.

Why did you and your research partners decide to build Distribuciones?

Flores: The idea was to take two scientific papers and make the results accessible to the general public — anyone who is interested in getting a better understanding of the data about inequality and the distribution of income in Latin America. Our studies cover 11 countries — that’s 80 percent of the Latin American population — over two decades. We are, of course, limited by the quality of the data. But we try to showcase every indicator possible, every definition, and explain them in simple, clear terms so that they can be understood not only by specialists. There are many sources and institutions that provide different numbers and use different definitions of income and units of analysis. It’s easy to get lost in all of that.

These two papers present new estimates about inequality. The official data that the government in Chile or Brazil or Mexico will give you are based on household surveys. But we know that household surveys have a very hard time capturing what happens at the very top of the distribution. They don’t capture capital incomes very well. That’s well documented and we also document it in our papers. What we’re doing here is taking the next step to say, Okay, let’s try to adjust and reach some consistency so that we can actually talk about the distribution of economic growth. If you’re missing half of the income, you cannot actually say what’s happening with economic growth.

In our studies, we use data from very different sources. We use household surveys for every country, and we have data from each country’s central bank or statistical institute that gathers data for the country’s national accounting. We use data from administrative sources, typically sources used either for social security or for personal income taxes. And we combine all of them in different ways to get a picture that’s closer to the macroeconomic numbers, to economic growth. Once we do all our adjustments, we compare the numbers to the national accounts to estimate the capital income that’s missing. We try to be very transparent in how we adjust, especially with incomes at the top and with capital income.

Our studies also bring comparability across countries. There are different pieces of information available for each country and they are not necessarily all comparable to each other, and we are trying to bridge that gap as well.

How do your findings differ from those in previous studies of the region?

Flores: For more than a decade, scholars published research saying that Latin America was the only region that had declining inequality in the world. Inequality was rising everywhere, but according to the official numbers, Latin America was the only region that was seeing inequality fall from very high levels. And that was very positive in a way, and people were happy about it and tried to explain it.

But the thing is, the decline in inequality was exclusive to labor incomes; capital incomes were underestimated. Now that we’ve adjusted for capital incomes and can see what’s been happening at the top, we see that the story is a bit more nuanced: inequality in the middle and bottom of the distribution — the bottom 99 percent — decreased while it increased at the very top. Most of the income in the top 1 percent is capital income; labor and other forms of income dominate in the middle and bottom of the distribution. And while labor income inequality was declining, capital income was going in the other direction. It was getting more and more concentrated.

Therefore, in the biggest economies — for instance, Brazil, Mexico, Colombia, and Chile — the level of inequality was actually much more stable, rather than declining. The diverging trends on the labor side and the capital side are balancing each other out. That’s the more nuanced story that we’re trying to tell, after years of research and pain and tears.

How can researchers — or the general public — access the data, and are there plans to add more?

Flores: Anyone can download the data. There’s so much of it, and it’s very detailed, but we took a lot of time to present it in a way that documents how all the variables are named, what they mean, and the organizations they come from. There are several options to download data in more aggregated ways or with a lot of detail.

I’ve been working on producing estimates on wealth inequality in Latin America based on administrative data, which is very rare. Those data are very difficult to access, but in the following year I should be able to access data in Chile. Some other people are working with Brazilian data. I’d love to make this a repository of wealth data for Latin America, showcased in ways that are accessible.

Read More: