The agricultural productivity gap is one of the main issues of developing economies. The methodologies used to assess these productivity levels are heavily reliant on assumptions over the production function, that, most of the time, do not take into account the heterogeneity of production and factor roles throughout the production process. In this paper, we will take a new approach to agricultural production through the lens of a task-based production model. The empirical strategy takes a novel perspective on market failure analysis by separating market dynamics at each production stage and having a holistic approach to all the production factors, taking the case of coffee farms in rural Colombia. The paper shows that labor and machine demands are biased by household characteristics, proving the existence of market failures in the economy and that such behaviors are more common in farms led by females due to credit market dynamics. Results also hint at the dependency of equilibrium factor allocation on other factors' markets and scales, which potentially causes a double productivity gap when facing a single market failure.
