Rafael Araujo
LightNews LightNews
banner
araujocrrafael.bsky.social
Rafael Araujo
@araujocrrafael.bsky.social
760 followers 600 following 32 posts
Assistant Prof. of Economics (FGV EESP)

rafaelcraraujo.github.io
Posts Replies Media Videos
araujocrrafael.bsky.social
We often think of transportation infrastructure as a key driver of economic growth. And that’s true. But it also has a big environmental footprint -- one that’s often underestimated.
June 17, 2025 at 8:21 PM
araujocrrafael.bsky.social
We derive bounds around the true linear effect of crop productivity on the outcome of interest. These bounds exhaust all the information contained in the first two moments of the data distribution and have intuitive closed-form solutions.
February 24, 2025 at 3:41 PM
araujocrrafael.bsky.social
We document the possibility of measurement error in the GAEZ variables. Their versions differ with respect to calibrated parameters and sources of data. These differences cause large changes in productivity for many crops and regions, suggesting the existence of mismeasurement.
There are two images here, on the left it shows a map of Brazil with the difference between versions 3 and 4 of FAO-GAEZ data for soybeans. On the right it shows a scatter plot of the data. There are two images here, on the left it shows a map of the South of the US with the difference between versions 3 and 4 of FAO-GAEZ data for cotton. On the right it shows a scatter plot of the data. There are two images here, on the left it shows a map of the Old World with the difference between versions 3 and 4 of FAO-GAEZ data for potato. On the right it shows a scatter plot of the data.
February 24, 2025 at 3:41 PM
araujocrrafael.bsky.social
The impact of crop productivity on economic outcomes is important in many fields. The most common measure of crop productivity is the FAO-GAEZ measure. However, this measure is based on a prediction model and is subject to measurement error.
this is a picture of the abstract of the paper. The FAO-GAEZ crop productivity data are widely used in Economics. However, the existence of measurement error is rarely recognized in the empirical literature. We propose a novel method to partially identify the effect of agricultural productivity, deriving bounds that allow for nonclassical measurement error by leveraging two proxies. These bounds exhaust all the information contained in the first two moments of the data. We reevaluate three influential studies, documenting that measurement error matters and that the impact of agricultural productivity on economic outcomes may be smaller than previously reported. Our methodology has broad applications in empirical research involving mismeasured variables.
February 24, 2025 at 3:41 PM
araujocrrafael.bsky.social
Natural sciences have studied these mechanisms for decades but economics can also bring new lessons. When farmers reoptimize their crop choices in response to climate shifts, economic outcomes shift too.
This figure shows two overlapped histograms showing the distribution of returns in scenarios with and without adaptation. The distribution of the scenario with adaptation is significantly to the right of the distribution of the scenario without adaptation.
November 11, 2024 at 7:07 PM
araujocrrafael.bsky.social
Take the Xingu Indigenous Territories (can you spot it?), removing Xingu’s protection status expands agricultural land but worsens the climate. Ignoring climate feedback would miss the bigger picture: 40% of the agricultural gains from deforestation are offset by climate feedback.
The figure shows a satellite image constrating a large green patch of forest (the Xingu) surrounded by large brown patches of agricultural land.
November 11, 2024 at 7:07 PM
araujocrrafael.bsky.social
Why does this matter? I apply the model to the Amazon, focusing on how protecting indigenous territories impacts agriculture. Flying rivers operate over long distances, meaning deforestation policies create a network of winners and losers
This figure shows two maps of the Brazilian state of Mato Grosso. One, on the left, shows the impact of Xingu's deforestation on forest cover. The scale varies from a zero effect to 10%+. On the right, the maps shows the impact of deforestation on the expected return. The scale varies from 0 to -20%-. The spatial pattern in both maps is that the center of the state, a region close to the Xingu, is the most affected region.
November 11, 2024 at 7:07 PM
araujocrrafael.bsky.social
“Does one exist?” I build a model where agricultural productivity depends on climate, and climate depends on deforestation. I show that equilibrium exists and show conditions for this thing to be a contraction.
This figure is a drawing showing three people dressed as spider-man pointing at each other. Each spider-man is labeled as "best response".
November 11, 2024 at 7:07 PM
araujocrrafael.bsky.social
In the tropics, agriculture drives deforestation, but it also depends on a stable climate, which is in part maintained by forests. This feedback creates a delicate equilibrium... if one exists.
The figure shows a satellite image contrasting a small patch of green forest with large patches of brown agricultural land.
November 11, 2024 at 7:07 PM
araujocrrafael.bsky.social
After many iterations my paper on endogenous climate in deforestation modeling is reborn!
(an amazing photo from the one and only Sebastião Salgado:)
The figure shows a massive rainfall event over the Amazon Rainforest. The picture, in black and white, highlights the composition of the clouds, the rainfall, and the forest.
November 11, 2024 at 7:07 PM