Using an equivalent variation approach, we estimate households’ willingness-to-pay for rural feeder roads in Ethiopia. The problem of endogenous road placement is addressed by a purposeful data collection process as the survey site was chosen in such a way that the primary difference between households in the otherwise homogeneous region is that transport costs to the same market differ substantially within the region (most remote households have to walk eight hours to reach the market). Using this quasi-experimental setting, we compare the economic behavior of households by remoteness, allowing us to estimate the benefits of having access to feeder roads for rural households. We find that the benefits of reducing transportation costs by 50 US Dollars per metric ton for the most remote households would result in benefits worth roughly 35 percent of household consumption and that a hypothetical gravel road built halfway through the survey site that lasts 10 years will have an internal rate of return that ranges from 12 to 34 percent, using conservative assumptions. The lower range of estimates allows for transport services that are restricted to intermediate means of transport such as donkey-drawn carts. These results suggest that investments in rural feeder roads are cost-effective ways to help reduce widespread poverty even in unfavorable settings where (a) small-scale farmers have low levels of marketed agricultural surplus, (b) nonfarm earnings opportunities are negligible, and (c) the provision of motorized transport services is not guaranteed.