Canonical urban models postulate transportation cost as a key element in determining urban spatial structure. This paper examines how road rationing policies impact the spatial distribution of households using rich micro data on housing transactions and resident demographics in Beijing. We find that Beijing’s road rationing policy significantly increased the demand for housing near subway stations as well as central business districts. The premium for proximity is stable in the periods prior to the driving restriction, but shifts significantly in the aftermath of the policy. The composition of households living close to subway stations and Beijing’s central business districts shifts toward wealthier households, consistent with theoretical predictions of the monocentric city model with income-stratified transit modes. Our findings suggest that city-wide road rationing policies can have the unintended consequence of limiting access to public transit for lower income individuals.