Brazil's $1B Farm Tech Investment Returned $17 for Every Dollar
Embrapa increased Brazilian agricultural productivity by 110% through targeted public R&D, generating $17 in benefits for every dollar spent.
Brazil's agricultural research corporation proves developing countries can escape the "technology mismatch trap" through strategic public investment in locally-appropriate innovation, transforming from food aid recipient to the world's third-largest agricultural exporter. A new NBER working paper, "Public R&D Meets Economic Development: Embrapa and Brazil's Agricultural Revolution" by Ariel Akerman, Jacob Moscona, Heitor S. Pellegrina, and Karthik Sastry, provides the first rigorous quantification of these returns.
By the numbers:
110% increase in Brazilian agricultural productivity from Embrapa's research
17-to-1 benefit-cost ratio (compared to 5-to-1 for U.S. public R&D)
39% of Brazil's total agricultural productivity growth since 1970
2,300 agricultural scientists employed by 2010
350+ crop varieties developed
200 international patents secured
The backstory: In 1973, Brazil faced a crisis. The country relied on foreign food aid while sitting on 2 million square kilometers of unused Cerrado savanna. Norman Borlaug said this land would never be productive due to acidic soils, aluminum toxicity, and nitrogen deficiency.
Brazil's response: Create Embrapa (Empresa Brasileira de Pesquisa Agropecuária) with a radical premise. Don't import foreign technology; develop solutions for Brazil's unique ecology.
How they did it:
Geographic spread over concentration: Embrapa established 40+ research centers across Brazil's six biomes, not in a single hub. Researchers in the Amazon studied Amazon problems; those in the Cerrado tackled acidic soils.
Focus on staples, not cash crops: While existing research targeted coffee and sugarcane exports, Embrapa prioritized beans, cassava, maize, rice, soy, and wheat.
Problem-based, not curiosity-driven: Scientists worked directly with farmers, developing agricultural liming to neutralize the Cerrado's acidic soils and creating the first soybean varieties for tropical latitudes.
Scale matched ambition: By 2010, Embrapa's $1.15 billion budget equaled 1% of Brazil's agricultural GDP, comparable to U.S. agricultural R&D spending as a share of sector output.
What the data shows: MIT and Princeton economists analyzed 35,000+ Brazilian agricultural scientists' careers and nine census rounds to quantify Embrapa's impact:
Embrapa researchers published 74% more articles about Brazilian biomes and 64% more about local pests than university researchers
Employment at Embrapa increased researcher productivity, especially in remote regions where it fully compensated for the productivity disadvantage of working outside major research hubs
Municipalities ecologically similar to new Embrapa centers saw agricultural output surge 30-40% within a decade
Effects concentrated on Embrapa's focus crops: productivity gains for targeted staples dwarfed those for other crops
Decentralization Dividend: Had Embrapa concentrated all resources in Brasília headquarters, productivity gains would have been just 70% instead of 110%. The benefit-cost ratio would drop from 17 to 11.
Operating from a single location in Brasília still beats the average return from other potential single centers by 50%, suggesting strategic headquarters placement.
Key mechanisms at work:
Technology mismatch matters: Research output's effectiveness decays rapidly with ecological distance. Technology developed for one biome often fails in another.
Local focus drives innovation: Researchers were 1.75x more likely to study their local biome and 2x more likely to research locally-present pests
Input intensification followed innovation: Fertilizer, seed, and chemical use surged in areas exposed to Embrapa research
Land use shifted: Cropland expanded while pasture declined as crop productivity improved relative to livestock
What didn't happen:
No crowding out of private research (Embrapa actually stimulated non-Embrapa agricultural research)
No increase in inequality (farm size inequality decreased in areas with greater Embrapa exposure)
No productivity penalty from geographic dispersion (Embrapa maintained high research output even in remote locations)
Historical parallel: Brazil's strategy echoes the Netherlands' post-WWII agricultural transformation, where Wageningen University anchored a government-led research and extension system that turned a war-ravaged nation into the world's second-largest agricultural exporter despite having 237 times less land than Brazil. Both cases demonstrate how coordinated public research institutions, farmer engagement, and technology development can revolutionize national agriculture.
Bottomline: Embrapa demonstrates that developing countries need not accept technological dependence. Strategic public R&D investment in locally-appropriate technology can drive massive productivity gains, but success requires both scale and geographic scope.
The model challenges Silicon Valley wisdom about innovation clustering. For technologies sensitive to local conditions (agriculture, medicine, climate solutions), spreading research infrastructure may generate higher returns than concentration.
With the low-hanging fruit harvested, returns to new Embrapa centers have declined from 60% productivity gains pre-1973 to 10% today. The organization now pivots toward climate resilience and sustainability rather than pure productivity growth.