We estimate the effect of democratic transitions on agricultural protection in a sample of 74 developing and developed countries, observed in the 1955-2005 period. We employ both differences-in-differences regressions and semi-parametric matching methods, exploiting the time series and cross-sectional variation in the data. Our semi-parametric matching estimates show that parametric methods might underestimate the true effect of democracy on agricultural protection. We find a strong increase in agricultural protection (reduce in taxation) after a country transition to democracy. Specifically a democratic transition increases agricultural protection by about 9 percent points. However, the effect is asymmetric as the effect of leaving democracy on protection is close to zero. The evidence supports the redistributive nature of democratic institutions toward the majority and, therefore, it is not inconsistent with the median voter model of political behaviour.