Association for Public Policy Analysis & Management

To combat traffic congestion and air pollution, many cities restrict vehicle ownership, but little is known about how these policies actually affect vehicle ownership, use, or travel time. Leveraging the randomization created by Beijing’s vehicle license plate lottery, we estimate the effects of the policy on travel behavior.

We establish three facts on the impact of vehicle ownership restrictions on travel behavior. First, the restrictions substantially decrease the number of cars on the road. By comparing car ownership rates of winners and losers, we find that these restrictions decreased the total stock of cars in Beijing by 15%, a larger amount than has been previously documented.

Second, vehicle ownership restrictions substantially reduce car usage for lottery losers. We estimate that each car approximately triples vehicle kilometers traveled and commute distance driven. Similarly, each car increases morning and evening car rush hour car usage by 33.2 percentage points and 28.7 percentage points.

Third, for women, each car increases total distance traveled by all forms of transportation by 32% and commute distance by 24%, implying a significant impact on commuting costs. For men, lottery losers substitute from buses and subways to cars in a nearly 1 to 1 ratio, so that every kilometer not driven by car is replaced by other forms of transportation.

These conclusions have important implications for localities considering vehicle ownership restriction policies. In combination, our three findings suggest that these policies have important environmental benefits in the form of substantial reductions in congestion, fuel consumption, and pollution emissions.