This paper integrates daytime and nighttime satellite imagery into a spatial general-equilibrium model to evaluate the returns to investments in new motorways. Our approach has particular value in developing-country settings in which spatially granular economic data are scarce. To demonstrate our method, we use publicly available imagery to evaluate India’s road construction projects in the early 2000s. Estimating the model and evaluating welfare impacts only requires remotely-sensed data. We find that India’s road investments improved aggregate welfare, particularly for the largest and smallest urban markets. Welfare gains were unevenly distributed across space, with the Golden Quadrilateral disproportionately benefiting large, already connected markets, while national highway upgrades delivered greater gains to smaller, more remote locations. More generally, within-district variation explains a large share of the overall spatial variance in welfare changes during this period, underscoring the value of high-resolution satellite imagery for capturing localized economic effects.