The cost of a free bus

Several cities have made public transport free. The results are more interesting than either side of the usual argument expects.

The case for is straightforward. Removing fares removes a barrier for people with the least money, speeds up boarding because nobody is paying, and saves the cost of collecting, counting and policing fares, which in some systems consumes a surprising share of the revenue it protects.

The evidence on car use is where the argument gets uncomfortable for supporters. In most cities that tried it, the new passengers came mainly from walking and cycling, not from driving. People who drive are usually choosing speed, door-to-door convenience and the ability to carry things. A bus that is free but still slower and less convenient does not change that calculation, because price was not the obstacle.

What consistently does move drivers onto buses is frequency and priority — a bus every six minutes in its own lane, so that it is actually quicker than the car. That costs money, and the fares that were abolished were helping to pay for it. A city therefore faces a real choice rather than a free lunch: the same budget can buy free travel on a mediocre service, or paid travel on a good one.

Neither answer is obviously right. Free travel is a direct transfer to people who need it; a faster network serves more people overall and takes more cars off the road. The useful discussion begins once both sides accept that the money can only be spent once.