Eliminating Uncertainty in Market Access: The Impact of New Bridges in Rural Nicaragua

We estimate the impact of new bridges in rural Nicaraguan villages facing seasonal floods that unpredictably eliminate access to outside markets. We collect detailed annual household surveys over three years and conduct weekly telephone followups with a subset of households for sixty-four weeks, both before and after construction. This information is collected in villages where bridges are built, and in comparable villages where no bridge was built for only engineering-related reasons. We find that bridges eliminate uncertainty in market access driven by floods: in absence of a bridge, household income falls significantly during floods. Bridges completely eliminate this effect. Compared to households in villages where no bridges are built, bridges cause substantial reallocation of activities between farming and wage work. There are also significant effects on agricultural choices: increased fertilizer spending, increased yields on farms, and lower crop storage. We develop a model of occupational choice and risky farm investment, and show that these results are a rational response to reduced market income risk from the bridges. We provide evidence that these results are not due to lower trade costs outside of flooding periods, nor due to looser collateral constraints.