2021 Projected Closure Statistics for selected crossings (projected by increasing trains per hour from 17 to 20 in peak direction plus counter peak direction) Source: Project 10,000

Victorian Labor’s big ticket transport idea for the November election is to remove 50 of “the most dangerous and congested level crossings on the metropolitan rail network” at an estimated cost of $5-6 billion.

Unlike other major Australian cities, Melbourne still has a very large number of level crossings in the suburbs, around 180, and there’s considerable pressure to eliminate all or most of them (e.g. see herehere and here).

Labor’s transport policy, Project 10,000, says crossings exacerbate traffic congestion, pose a safety threat for pedestrians and drivers, and impose a limit on the number of trains that can operate on a line.

Boom gates at many of Melbourne’s most congested level crossings are already down for anywhere between 30-40 minutes in the hour during peak times.

Labor says its promise is estimated to cost $5 – 6 billion. It’s already nominated 40 crossings for grade separation and says a further ten will be announced before the election.

It says funding will come from the sale or lease of the Port of Melbourne and from creating development opportunities associated with grade separating level crossings (1).

There’s no mention of timing in Project 10,000, but the media has reported Labor’s overall timetable for removing the 50 level crossings is eight years (see here and here).

Grade separation is a politically appealing idea because it benefits both motorists and public transport users (e.g. see Does removing level crossings benefit public transport users?).

The sort of high train frequencies necessary to create a truly “turn up and go” public transport system simply aren’t politically feasible if motorists – who account for the great majority of all peak hour trips in Melbourne – only have a short window at crossings. (2)

Eliminating level crossings also spreads the works across a very large area of Melbourne; most motorists can see how they might benefit from eliminating one or more level crossings they use regularly.

But as always, it’s never that easy; there are a number of issues to consider.

First, the Government disputes the cost, arguing the program will cost $8.5 billion. Project 10,000 assumes an average cost of $100 – 120 million per crossing, but it’s promising to prioritise the most congested crossings in Melbourne. These tend to be more costly; up to $150-200 million each. (3)

Second, the funding is planned to come from selling/leasing the Port of Melbourne. However it’s not clear yet if that will deliver the required revenue and/or whether it will be available within a time frame consistent with Labor’s promise. There’s also the question of how using the revenue for another purpose will affect the funding of a new port.

Third, Project 10,000 doesn’t explain how the works will be sequenced. In order to enable higher train frequencies, all crossings have to be removed from a line. The danger, though, is that works will be prioritised by political expediency rather than in accordance with operational priorities for rail; if so, that could primarily benefit motorists rather than train users.

Fourth, Labor’s list is largely based on the RACV’s priority list for level crossing eliminations. It’s unclear what degree of analysis has gone into the RACV’s exercise and what relative weighting it gives to different stakeholders e.g. public transport users vs motorists.

Fifth, the impact of grade separations on local amenity, pedestrian accessibility and established retail precincts can be substantial in a relatively flat city like Melbourne.

This can be addressed in part by putting rail under road, but that’s very expensive. A recent study suggests the optimal way to handle the many trade-offs is to elevate rail over road, although that too will likely cost more than Labor’s assuming.

Finally – and most importantly – the traffic congestion benefit of grade separations has to be assessed carefully. There’s an argument that it’s illusory; that grade separations will simply move peak hour congestion on major arterial roads to the next major choke point, usually an intersection. This is a key reason why road authorities tend to argue for freeways.

Labor’s plan might seem like a “no-brainer” but we don’t really know if the funding can be provided within a reasonable time frame and/or whether it’ll be compromised by local politics. Moreover, like the Government’s planned East West Link motorway, we don’t know if the benefits will materialise to the extent claimed or if they’ll exceed the costs.

Constructing some of these grade separations might well prove to be a good idea to improve train frequencies, but it’s a pity our political culture encourages politicians to commit huge sums on the basis of relatively skimpy evidence. (4)


  1. The financial contribution from “value capture” is emphasised in Project 10,000 but is likely to be quite modest relative to the cost.

  2. Motorists also include bus and tram users.

  3. The Opposition appears to have inadvertently confirmed the cost would be higher i.e. $8 billion for 48 grade separations.

  4. There are other possible ways to increase train capacity in the peak hour besides higher frequencies e.g. longer trains and platforms, fewer seats. Investment in improved signalling can also reduce the time that boom gates need to be down.