We show that isolated capital cities are robustly associated with greater levels of corruption across US states, in line with the view that this isolation reduces accountability, and in contrast with the alternative hypothesis that it might forestall political capture. We then provide direct evidence that the spatial distribution of population relative to the capital affects different accountability mechanisms over state politics: newspaper coverage, voter knowledge and information, and turnout. We also find evidence against the capture hypothesis: isolated capitals are associated with more money in state-level campaigns. Finally, we show that isolation is linked with worse public good provision.