Counting the cost of the Sydney Metro

The latest warning over the impacts of the Sydney Metro currently under construction on the rest of the city’s rail network doesn’t say anything particularly new. What makes it special however is that it was written not by committed lobby groups or disgruntled academics but rather by some of the state’s most senior and respected former transport executives. More importantly it lays bare the mounting costs of the obsession of successive state governments with the metro concept.

According to the Fairfax article the analysis (obtained by Labor under freedom-of-information laws) was actually completed as a submission in July 2015 by a group including Ron Christie, former State Rail Authority CEO, John Brew, a former State Rail chief executive, Bob O’Loughlin, a former State Rail director of rail safety and operations and Dick Day, a former head of planning and timetables at RailCorp. According to the article their submission warns that the metro plans “will result in ‘degradation of the robustness and reliability’ of Sydney’s existing heavy rail network, and ‘ultimately lead to the total network becoming gridlocked and unworkable’”.

This is because the conversion of the Bankstown line for metro operation will diminish the network’s capacity, removing a “relief valve for the network” and leaving “no escape route”. In addition, the authors claim that the line “will do nothing to relieve” the bottleneck at Strathfield where Western and Northern line trains merge.

They authors state that the $17 billion cost of the metro would have been money better spent on improving the existing rail network and that there is likely to a “voter backlash” to the new single-deck trains which will provide significantly fewer seats than double-deck trains (and fewer proportionally than any of the other Australasian CBD metro projects currently under construction).Variations of these arguments have been made virtually since the announcement that the link would be built as a stand-alone metro. The case was most eloquently put by transport expert Sandy Thomas in his 2013 ‘Fixing’ the Trains in Sydney: 1855 Revisited article and while subsequent events have caught up with some of his analysis it remains an essential introduction.

I won’t rehash the history of the Metro project (I covered it in an article earlier this year) but it’s useful to summarise the context that Thomas provides. As he notes, one of the final announcements of Bob Carr’s premiership in 2005 was that the NSW government would build three new conventional passenger rail lines in Sydney – the South West Rail Link (SWRL) from Glenfield to Leppington by 2012, a North West Rail Link (NWRL) from Epping to Rouse Hill by 2017 and a new “Harbour Rail Link” between St Leonards and Redfern, adding a new underground line through the CBD, also by 2017. These new lines were to be supported by additional tracks on existing lines between Chatswood and St Leonards and between Redfern and Campbelltown.

What might have been: the proposed Metropolitan Rail Expansion Program, 2005 (source: Sandy Thomas via Transport Sydney website)

The plan, in development for a decade (led by Dick Day), was called the “Metropolitan Rail Expansion Program” (MREP). Thomas observes that the MREP “was intended to cater for more frequent, faster and longer (ten-carriage) double deck trains, including long-distance expresses, on a new, separate operational “sector” of the CityRail network…. boosting the capacity of CityRail/Sydney Trains lines through the CBD by almost 30% and, by permitting much more efficient use of the capacity of these and other lines, almost doubling the total patronage capacity of Sydney’s CityRail/Sydney Trains system”.

In effect the MREP would have become like Paris’s RER lines, providing both outer-suburban services and a limited-stop express route through the heart of the CBD. Unfortunately, despite Labor remaining in power for another six years, only the SWRL section was commenced. Thomas describes in depressing detail the internal war during this period within government between the supporters of the MREP and their pro-metro opponents (and not to mention those who were opposed to any public transport infrastructure investment at all).

A lot of the drive for the metro and against heavy rail was ideological and even at times irrational. There was a perception that RailCorp was incompetent at major project management, the unions were running the place, double deck trains were old technology and, conversely, that privately-run metros were the new shiny things of the future.

For a time the metro faction won out, with the Labor government announcing a succession of metro proposals as alternatives to the MREP. For a variety of reasons none of these made it off the drawing board and when the Coalition swept to power in the 2011 election with a commitment to construct the NWRL as a heavy rail link it looked like the matter had been settled decisively. That was until the government’s about-face in 2012 with the announcement that it would build the NWRL as a metro and extend the line under the harbour and to Sydney’s south, taking over the whole of the Bankstown line as well as providing a link to Hurstville.

This effectively replaced the MREP with what is now called the Sydney Metro (later the link to Hurstville was dropped and the Bankstown line takeover shortened to Bankstown itself). The Government’s public rationale for the switch in 2012 rested largely on two claims – that the metro would provide much greater capacity than a double deck line and that it would be cheaper to build.

Sydney Metro currently under construction (source: transport for NSW)

The first claim has largely been rebutted by Thomas and others. I don’t want to go into the complexities of peak and crush capacities of double deck trains and metros, but if the MREP had been built as planned and equipped with modern signalling, a frequency of 24 trains per hour would have been feasible. The line would have had a crush capacity of over 40,000 in each direction, not much less than the metro line will have at 30 trains an hour (both scenarios assume eight-car trains). There is one big difference – over half the passengers on a double-deck train would have been seated, compared to less than a third on the metro trains.

I want to concentrate on the second claim. Much of the cost-saving argument in 2012 was made by then Transport Minister (now Premier) Gladys Berejiklian, based on the assertion that the metro would require lighter viaducts and smaller-diameter, more-manoeuvrable tunnels than double-deck trains.

What was not immediately clear at the time was the cascading set of consequences resulting from the metro decision. Because it is completely incompatible with the current network, any direct connection to it must be “upgraded” to metro standard, which means it cannot be easily integrated with the existing lines at either end in the same way envisaged under the MREP.

In dollar terms, this also means that any savings on new construction west of Epping and from Chatswood to Sydenham will be more than off-set by the cost of the following:

  • The Epping-Chatswood Rail Link (ECRL) metro conversion, which would not have been required for the original MREP plan. This is a particular waste of money as the line is less than 10 years old and is virtually being gifted to the private operator.
  • The extension of the metro tunnel south of the CBD from Redfern (where it terminated under the MREP) to Sydenham to provide a direct connection to the Bankstown line. While the MREP did propose additional tracks in this section, these could have been added gradually and at much less expense than tunnelling.
  • The Bankstown line metro conversion. An important component of the MREP was that by providing a new path through the CBD for trains from the southwest it would free up capacity of the City Circle for trains from other lines, particularly the west. Because it is clearly not viable to convert the whole of the existing southwest line to metro to fulfil this role, a corridor closer to the CBD (and conveniently ripe for development) – the Bankstown line – was selected instead. As noted earlier, connection to the metro means conversion. However, this is looking more and more like a complete rebuild of the entire Bankstown line, stations, bridges and all, to metro standard – and, like the ECRL, it will be effectively another gift to the metro’s private operator.

Then as the former senior transport official’s submission points out there are the other less direct costs, mainly related to the impact on the rest of the existing rail network. Specifically, while the metro may indirectly increase capacity in the CBD for the existing network by providing an alternative pathway for Bankstown line trains (plus a limited number from the Northern line), it will actually have the opposite effect elsewhere.

The key issue identified both in the submission and by Sandy Thomas is that the metro will exacerbate rather than relieve the bottleneck between Strathfield and the CBD which is used by Western and Northern line trains. The latter will no longer be able to use the link between Epping and Chatswood as an alternative after its conversion to metro; instead they will all have to revert to the pre-ECRL pattern of travelling to the city via Strathfield and this crowded section of track, which amounts to a significant loss of capacity on this line. Another issue identified by Thomas is that unlike the MREP proposal the metro does nothing to directly increase capacity for the Main South and East Hills lines, one of Sydney’s most congested corridors.

And even with increased capacity in the CBD the metro will have a constraining effect there. Currently there are three double-track corridors through the Sydney CBD available for double-deck trains – the City Circle, the North Shore Line and the Eastern Suburbs Railway. Under the MREP the new CBD line would have added a fourth; while the metro will add another CBD line it won’t be available to the existing system. Given the government’s enthusiasm for metros and the specific plans for a metro to Parramatta it looks increasingly likely that the current rail network will be permanently locked into its current level of capacity in the CBD.

This has major ramifications for future development of the whole system. Ultimately it will be only the metro corridor currently under construction and the proposed Sydney West Metro to Parramatta that will have spare capacity in the CBD, either to accommodate potential new metro lines (for example, to the northern beaches) or for the conversion of additional existing rail lines.

Given the state government’s enthusiasm for metros, driverless trains and privatisation this is probably not a surprising result. Whether the cascade of decisions resulting from the initial metro announcement were all originally part of the government’s plan to sectorise and privatise the existing rail network is now a moot question; clearly however the government intends to make use of the consequences to drive further privatisation as well as to provide a catalyst for major redevelopment along the metro corridors.

This is not an argument against metros per se – indeed a metro line to Parramatta makes a lot of sense (as would a metro line to the south east instead of the light rail, but that’s another story) – but it does demonstrate how ideology and even fashion can drive enduring, expensive and suboptimal results in terms of efficient public transport infrastructure investment.

This entry was posted in Infrastructure, Public Transport, Sydney metro area, Transport, Western Sydney and tagged , , , , , , , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s