Guatemala Ex-Ante Impact Model

An ex-ante model of the nutritional and economic impacts of biofortified maize in Guatemala, developed for New Seed.

In Guatemala, maize is central to daily life, and chronic child malnutrition remains high.

New Seed develops high-yielding, climate-resilient biofortified maize seed and works with seed companies, governments and other partners to make it accessible to farmers. As adoption grows, biofortified grain enters local and national markets, so it can reach consumers who never grow biofortified maize themselves.

This model estimates what that could mean for child nutrition and farmer livelihoods at scale.

Rationale

Why an ex-ante model

Measuring the full impact of an intervention only after it has reached national scale would mean waiting years to answer questions that matter now: where biofortified maize could have the greatest impact, how different levels of farmer adoption might affect child nutrition and farmer income, and how improved seed varieties or changes in seed subsidy levels could alter those outcomes.

An ex-ante model does not predict a single future. It provides a structured way to examine how projected outcomes change under different assumptions, and makes those assumptions explicit.

Scope

What the model covers

The model combines Guatemalan national survey data with evidence from New Seed and published research, following the pathway from seed to impact:

  1. which farmers are likely to adopt biofortified maize;
  2. how adoption may affect maize production and farmer income;
  3. how much biofortified maize could enter the food system;
  4. how this could change children’s intake of iron, zinc, and quality protein;
  5. and how modeled changes in nutrient intake may translate into changes in child growth and stunting.

Scenarios vary by geographic area, seed variety, seed subsidy level, and the share of national maize production that is biofortified.

Where to go next

Access

Interactive

Interactive results

Explore the model dynamically and see how the projected outcomes change across different scenarios.

Compare geographic areas, seed varieties, subsidy levels, and percentages of biofortified maize production, and examine their estimated effects on nutrition, child growth, and farmer income across Guatemala.

Documentation

Methodology and documentation

See where every number comes from.

The complete analytical process is documented from the original data sources to the final national simulation, including data preparation, assumptions, statistical models, validation steps, sensitivity analyses, and limitations.

The methodology and code are public so that the results can be examined, scrutinized, and reproduced.

What comes next

Beyond Guatemala

Guatemala is where New Seed’s work began. The work has now expanded to El Salvador, with plans to expand into Honduras next. New Seed is also working with partners in East Africa, with a stated goal of reaching 100 million people in Sub-Saharan Africa by 2035.

The evidence built in Guatemala informs this expansion: it indicates what works, what it costs, and where the largest opportunities may lie. Extending this model to another country requires re-deriving the analysis with that country’s own national datasets.

For information about supporting New Seed’s biofortification program, visit semillanueva.org/donate (opens in a new tab) or contact info@semillanueva.org.

Funding

Project support

This project was funded by grants to New Seed from Light a Single Candle, the USAID Development Innovation Ventures (DIV) Program, and Project Resource Optimization (PRO).

With special thanks to Mulago Foundation, Cartier Philanthropy, and Dovetail.

About

About the model developers

The model was developed for New Seed by Azahar Data Insights (opens in a new tab), a biostatistics and data science consultancy working in healthcare and life sciences. The six-module analytical framework behind it was built in R, from raw survey files to the interactive results: Quarto for the analytical pipeline and the methodology site, Shiny for the application.

The work involved deriving individual nutrient intakes from household consumption data, transferring those estimates between national surveys, modeling farmer adoption and income effects by segment, and building the simulation layer that produces the scenarios. The framework is documented for external audit, and the code is openly licensed.

Framework v4.4.0. Results build: 2026-09-15. Source code and methodology: github.com/Azahar-Data-Insights/newseed-biofortification-guatemala-public (opens in a new tab). DOI: 10.5281/zenodo.22940768 (opens in a new tab).