Tallinn’s fare-free rapid public transport network

There was lots of public commentary last week on the ill-advised idea currently under consideration by Victoria’s Legislative Council of expanding Melbourne’s CBD free tram zone. The consensus of critics is that it would exacerbate existing overcrowding of trams in the city centre, replace even more walking trips, and continue to provide no financial benefit to the great majority of commuters.

Largely overlooked in the argument was the parallel proposal to make public transport fare-free for students and seniors. This idea would exempt 60% of the state’s population from paying fares on all modes, so it was no surprise the Lord Mayor of Melbourne, Sally Capp, was prompted to look at the bigger picture.

“Wouldn’t it be great”, she said on radio 3AW, “if our entire public transport system was free…”

This is one of those ideas that keeps cropping up. Given recent initiatives like the CBD free tram zone and the effective abolition of the zonal fare structure, it’s conceivable in the near future that a political party will find the temptation to promise “free” public transport irresistible.

But is it a sensible idea? As Cats et al observe, the research suggests that fare elasticity is strongly asymmetric:

Passenger demand decreases in response to an increase in price but the effect associated with a price reduction is insignificant.

Very few places in the world have zero public transport fares and the few that do are very small. However there’s one recent entrant to the ranks, Tallinn, which is comparatively large (population 434,000). This Estonian city introduced free public transport in 2013.

But according to a World Economic Forum news report, the results are less than impressive. A before-and-after study by Dutch and Swedish researchers found that while public transport use had increased by 14% after almost one year, 40% of the increase came at the expense of active transport i.e. mostly walking.

A one-off 14% increase in patronage over 12 months isn’t especially large. For example, it’s the same as the organic growth in passenger boardings on Sydney’s trains, buses and ferries over the two years to April 2019 (i.e. before NW Metro opened).

Or look at it another way; if the number of trips Melbourne residents make by public transport increased by 14% – drawn half-and-half from cars and active transport – public transport’s share of all weekday trips would rise from 9% to 10%, driving’s would fall from 73% to 72%, and walking/cycling from 19% to 18%.

Of course, Melbourne and Tallinn aren’t alike. Apart from the big difference in population, public transport’s mode share was 56% in Tallinn before the new policy, car’s share was 32%, and walking’s was 12%. That’s a far cry from Melbourne, where public transport accounts for only 9% of trips, cars for a whopping 73%, and active transport for 19%.

So perhaps the increase in patronage from doing away with fares might be smaller, or greater, in Melbourne. If the jump were as big as 30% (say), the mode share of public transport for weekday trips by residents would increase from 9% to 11%, cars would stay at 72%, and walking/cycling would decline from 19% to 17%.

Those numbers are hardly the sort of game-changing outcome that might justify a radical policy like foregoing all fare revenue. The probable cost to the state budget in Melbourne would be roughly $750 million annually after netting out collection and enforcement costs.

That’s a significant outlay that would have to be taken from elsewhere in the state budget. As an indication of its scale, $750 million p.a. would be enough to double the size of Melbourne’s tram fleet over 10 years by buying outright 500 new E-class triple carriage trams. It could service over $30 billion in government debt – enough to double the size of the tram network.

It’s not an amount that should be taken lightly. The modest pay-off relative to cost isn’t the only likely drawback.

Fare-free travel might also make living further from the city centre a more attractive option. It would probably only provide a mild boost to outer suburban sprawl, but a much larger incentive for regional sprawl, given that fares for trips from regional centres are a significant outlay.

The equity effects of eliminating fares is complex. It would benefit low income groups and non-workers (these were the main beneficiaries in Tallinn), but it would also further subsidise city-centre workers, as well as students with good lifetime income prospects.

While a 14% increase in public transport trips would have a small effect on mode share, it would amount to an additional 160,000 trips per weekday in Melbourne. That’s more than double the daily patronage on the Hurstbridge and Mernda lines combined.

It would have a big impact in terms of crowding, with a high probability of downstream negative impacts on reliability and punctuality. It would necessitate significant additional capital and operating expenditure by government, with inevitable lags.

Any additional patronage might be thought of as desirable, but it’s not obvious that replacing walking, or incentivising existing public transport users to make additional trips, is justified at such a high cost.

Rather than abolish fares, the counter argument is fare revenue would have a bigger impact on patronage if it were used to make public transport more useful, especially if it could attract travellers away from driving.

If the government were contemplating effectively “spending” an additional $750 million per year on public transport, it would likely generate more trips by existing users, as well as draw more travellers from cars, if it were used to make public transport more attractive, for example by providing higher frequencies, greater reliability, and better connections.

Generating mode substitution is very hard in a city like Melbourne because cars are attractive in the sort of low density, dispersed setting that characterises the great bulk of the metropolitan area. That $750 million p.a. would have a much bigger impact if it were spent in part or whole on actions to make driving less competitive with public transport, for example by making cars slower or more costly to operate.

Road pricing, as recommended by Infrastructure Victoria, is one option. It would be far more difficult politically than spending (or foregoing) more money on public transport, but it’s a necessary condition for making serious inroads on the 73% of all trips (or 90% of motorised trips) Melbourne residents currently make by car.

Fare-free travel sounds appealing, but it probably wouldn’t generate growth in patronage, or more importantly mode shift, commensurate with the cost. The revenue would likely have a bigger impact if it were spent on making public transport better and driving less attractive.

And as for Melbourne’s CBD “free” (i.e. 100% subsidised) tram zone, what a foolish idea. Rather than waste time contemplating the political benefit of expanding it, the Andrews government and politicians of all stripes should acknowledge the blunder and walk away from it.