Brendan Keenan: Put all cards on table, then make decision on Metro deal
THE way the late Seamus Brennan told it, as Minister for Transport he read in the newspapers about the new underground metro in Madrid, and how it appeared to have cost a fraction of the plan to build one in Dublin. Having failed to get convincing explanations, he invited the Spanish engineers over to explain how it was done.
As we know, it never got done. Officials tut-tutted that the minister's account was a bit simplistic. And Mr Brennan's reputation, though admirable in many ways, suffered from his revelation that early estimates for the Luas were "back of the envelope."
Twenty years later, Dublin got to the stage of having preferred bidders for the first bit of the Metro, the North line to the airport, before it was deferred indefinitely by the present minister, Leo Varadkar.
Of course, Madrid has a real metro – 290 kilometres of track and 300 stations. It would be more accurate to call the Dublin project a track. Admittedly, Madrid started 90 years ago, but it is still expanding.
The bit that interested Mr Brennan was Metro Sur, a huge expansion which included 41 km of tunnel and 28 new stations. Incredibly, the whole thing was competed in less than three years. The cost of €85 million per kilometre astonished more than Mr Brennan.
At up to €3bn – the estimate from another minister, Eamon Ryan – Metro North would work out at €180 million per kilometre. To judge from information from different projects around the world, this would be just about the most expensive recorded anywhere.
Now along comes Cormac Rabbitt, who has got the ear of Dublin's former Lord Mayor, Andrew Montague, and other councillors. Mr Rabbitt, an engineer by training who once worked for the Dublin local authority, makes a habit of telling government ministers how their prestige transport projects could be built an awful lot cheaper.
A persistent habit. He presented ideas for a cheaper Luas to Mary O'Rourke as Minister for Public Enterprise. She was intrigued by what he had to say, but felt the plan was too far advanced to change.
In 2003 he outlined bargain basement prices for a metro to the Oireachtas committee on transport infrastructure and now he is back, offering them again for the deferred Metro North and the even more deferred West.
He explained to the councillors how Spanish practices, like cutting one tunnel instead of twin bores, and using the mysterious Phoenix Park tunnel, could deliver more metro for half the cost.
(The park tunnel is mysterious because, depending on which project is under consideration, it is either a useless relic or the answer to Dublin's public transport problems. Both positions have been held at different times by the same state agencies).
This column is not getting into the engineering business, and is well aware of the different soil, planning laws, environmental regulation and working hours in Madrid, but one does not have to be an engineering expert to have one's doubts about what has been going on here.
They are not helped by a proliferation of quangos. The Rail Procurement Agency was established to, well, procure new rail projects, which had been the remit of CIE. Having procured them, it held on to them and became quite a beastie, with a staff of almost 300 and a payroll of €22 million, although it is now downsizing.
It has been joined by the National Transport Authority, which absorbed the Dublin Transport Authority, which lacked the authority to get things done. The NTA is now the contracting entity, with the RPA as its agent. Whatever all that means.
Mr Rabbitt's suggestions are not just about boring tunnels. He would establish a consortium which would build and operate the system, supported by government grants – public transport never pays its way.
This is a step beyond the Luas arrangements, where a private company runs the trams under franchise (currently up for renewal) but the thing belongs to the state, in the shape of the RPA. Or maybe the NTA.
Private ownership would not help Mr Rabbitt's case politically. The reaction of politicians to his presentations has been that they would like to think he was right, but cannot quite believe it. That would be pretty much the reaction of most people. The trouble is that there seems no way of knowing.
A new book from the Economic and Social Research Institute* analyses ways in which we might know more about how government arrives at these decisions; and perhaps how government might know better how it arrives at its own decisions.
Whether or not the Spaniards have found the secret, the book points out that public transport projects everywhere seem to run into problems of evaluation and cost control. The average cost overruns budget for rail projects was a whopping 45pc. The traffic forecasts which justify plans are on average 23pc too high.
Paradoxically, this "optimism bias" tends to favour more expensive solutions, which may explain Mr Rabbitt's lack of success. International research suggests it is easier to get government funds for glamorous projects, rather than cheaper, less visible ones.
The choice of trams over dedicated busways in Dublin has always looked like one such example.
As the book's editors suggest, serious examination of alternatives should be part of the planning, but rarely is.
Even as the Irish public investment programme disappears down the black hole of current spending, plans are afoot to improve analysis and implementation of projects. There will be yet another agency, the Office of Government Procurement, which will "spearhead" the reform of procurement in the public service. We have already been promised the elimination of cost overruns through better planning and contracts which transfer as much risk as possible to contractors and consultants.
The book, which is aimed at a specialist audience and covers the whole range of policy, not just transport, reckons the existing Irish rules for infrastructure are too optimistic and that project planning is not wide enough, or open enough.
Lack of transparency is the reason we are all left wondering whether Mr Rabbitt's remarkable projections could possibly be true, and worrying about favouritism, or worse, in the awarding of contracts.
"It is difficult to have confidence in any analysis that is not made public, or of which only selected aspects are made public," the authors say.
"It is imperative that all evaluations should be published in full in order to allow public scrutiny." Imperative, yes; likely, no.
*Using Evidence to Inform Policy. Edited by Pete Lunn and Frances Ruane. Gill & Macmillan €39.99