The economic folly of free buses

ZOHRAN MAMDANI, New York's leftie mayor, campaigned on the promise to make them free. They already are for residents of Tallinn, Estonia’s capital, and in several French regions. In Germany they were at one point so heavily subsidised that they might as well have been. Voters love them—and so do politicians, especially populist ones. Free buses, or extremely cheap public transport in general, are all the rage. They are also terrible economics.

Advocates of the idea proffer three main arguments. First, fares discourage use, raising the system’s cost per passenger. This in turn may lead to underinvestment and, as a consequence, a sparse and low-frequency service that is not as economically efficient as it could be. Lowering the cost barrier encourages ridership, a denser network and a better service for everyone. Second, free transport takes cars off the road. Because motorists do not pay the full social cost of congestion, pollution and other “externalities”, they do too much motoring. A zero-cost alternative makes even socially underpriced car journeys less attractive, cutting traffic and accidents or making room for trees and playgrounds. Third, low or no fares make it easier for poorer households and non-motorists (including the young, elderly and disabled) to participate in public life. They also promote physical activity (ie, walking or, if you see your bus about to leave the stop, sprinting). That is free public transport as social policy.

All three arguments have merit. But none of them justifies giving public transport away for nothing or nearly nothing. Take economic efficiency. For this to rise, subsidised schemes would have to increase ridership substantially. Yet evidence on the ground suggests that their effect on passenger numbers is at best marginal. In other words, public transport turns out to have what economists call low price elasticity of demand.

In Tallinn, where public transport has been free since three-quarters of voters backed the idea in a referendum in 2012, the related rise in trips was just over 1%. In Santiago, Chile’s capital, a randomised trial in which some commuters got a two-week pass found a bump of 23% but only during off-peak hours. The German scheme, which charged €9 ($9.50 at the time) a month to travel on any public transport anywhere in the country when it was introduced in 2022, looks effective only by comparison. A new study in the Journal of Public Economics finds that train trips longer than 30km rose by 35%. But the largest increases were recorded on weekends and to rural tourist hotspots, which fills spare capacity but does little to encourage network expansion. Another new study using mobile-phone tracking data, by Daniel Horcher of the Corvinus University of Budapest and co-authors, shows that the initial pick-up in leisure trips faded quickly.

Ultra-low fares can also have bad effects. In Germany they led to 30% more train delays where the already stretched system struggled to cope with more travellers. Proponents argue that subsidies must go hand-in-hand with service expansion to meet the additional demand; this has happened around Lens, where ridership jumped by 40%. However, expanding services may do a better job at attracting more passengers than increasing passenger numbers with free rides does at expanding services. It is also cheaper simply to put on more buses. Some counties in Estonia limited the scheme to the young and old. They spent the savings on service expansion, with the result of comparable growth in ridership at lower cost per additional passenger.

Evidence is also patchy when it comes to deterring car use. A review of more than 150 related papers found little substitution away from driving. The German €9 ticket may have reduced car traffic by just 1-5%, and even that was not permanent. The effect could be larger in Luxembourg, where public transport has been free since 2020. A new working paper comparing the principality with similar European regions found that car-linked carbon emissions had declined by 6% and nitrogen-oxide pollution by 10%. But the uncertainty around these estimates was high, and it is in any case a costly way to protect the climate. According to one German analysis, the €9 programme reduced carbon emissions at a cost of €6,140 per tonne, a comically high figure next to the €85 per tonne in the European Union’s emissions-trading scheme.

A better way to cut driving is to price road use directly. As cities from Stockholm to Singapore show, congestion pricing works. So does regulating air pollution. London's “ultra-low emissions zone” now covers the entire city, not just the centre. According to one new study, this has helped lower the share of children with impaired lung function from 14% to 9%.

Is free public transport at least good social policy? Since poorer people tend to use more public transport, they stand to benefit most. But low-earners often cannot take full advantage of these benefits because many of them start work early or finish late, when services are thin. And the benefits also accrue to high-earners who could easily cough up for a ride. London’s free public-transport passes for anyone over 60 travelling after 9am costs the city £220 ($298) each year per person, even if that person is a loaded City banker. As for public health, free transport can be both a boon (it led to more walking in Chile) and a bane (in Estonia and Germany, it came at the expense of walking and cycling).

There is no such thing as a free bus

Public transport is costly, even when it is free. Germany has raised the price of its Deutschlandticket to €63 a month and still tops it up with about €3bn a year. This money has to come from somewhere: higher taxes, lower spending elsewhere or more debt (and so higher taxes or lower spending in the future). Mr Mamdani is already talking less about free buses and more about faster ones. His fellow progressives should do the same. ■