You are here

OPINION

London's £17.6 billion  supertrain will be worth the wait

Crossrail's cost hike and delayed opening may have raised hackles, but this outcome wasn't really all that unexpected

WHEN they're not busy bemoaning a shambolic Brexit, Britons now have another embarrassment to lament. London's 17.6 billion pound (S$30.5 billion) Crossrail project was meant to open this month, but it won't. 

On Monday, London's authorities confirmed that the railway connecting Heathrow Airport in the west with Canary Wharf in the east will need up to £2 billion more funding (lifting  it from the £15 billion or so initially expected). An already revised Autumn 2019 opening date might not be met.

Critics have branded Crossrail's missteps as a "catastrophe" and  a "shambles". The project's chairman, Sir Terry Morgan, has stepped down. But is Europe's biggest construction project really such a big failure? So far, it's not even close.

Railways, airports, bridges and concert halls, often costing upward of a billion dollars, are so frequently tardy and financially disappointing that academics say they obey an iron law: "Over budget, over time, under benefits, over and over again," is how Oxford academic Bent Flyvbjerg puts it. With some justification. The probability of cost overruns on these ventures is about 90 per cent. That doesn't mean we should just give up on major building plans, though.

sentifi.com

Market voices on:

Of course, big infrastructure projects provide plenty of fodder for the cynics. Modern regulatory and planning systems mean slowness is pre-ordained, while construction can be hugely expensive partly because of low productivity and rent-seeking. Rail projects typically cost 40 per cent more than budgeted, according to Prof Flyvbjerg, while dams tend to cost about twice as much as planned. Sometimes even that feels like a bargain.

Consider Berlin's new international airport - BER - which was meant to open in 2012 but won't start handling passengers until 2020. Costs first estimated at 2 billion euros have ballooned to more than 7 billion euros (S$11 billion). Germany, whose international reputation for precision and efficiency is sometimes baffling, shows that Britain's struggles with expensive infrastructure are hardly unique.

A list of recent German white elephants include the Hamburg Elbphilharmonie concert hall, Stuttgart railway station and the BND intelligence agency headquarters. In the US, Boston's Big Dig tunnel programme surely merits a place in the mega-project hall of infamy.

By comparison, Crossrail's problems - the ones we know about at least - appear manageable. So far the total budget has risen 11 per cent compared to the target agreed a decade ago. That's no small feat in view of its engineering complexity and the number of stakeholders involved. Since tunnelling started in 2012, eight giant boring machines have dug 42 km of new rail tunnels up to 40 meters below the capital. Engineers have also almost completed 10 new stations in one of Europe's mostly densely populated areas, yet kept business disruption to a minimum.

Digging the tunnel turned out to be the easy bit. Getting trains to recognise three different signalling systems along the route has been more trying, and testing hasn't been completed. That's not so unusual. A mega-project's "uniqueness" is often what causes it to run into trouble because engineers can't copy what they've done before. Seemingly run-of-the-mill details can end up derailing the whole thing. At BER, a poorly designed fire safety system was the chief problem.

Smaller, less sexy public works are often safer financially than those that win architecture prizes or set records. The same's true for the private sector, where Airbus  and Boeing have learnt that incremental innovation - giving existing aircraft new engines - is less perilous than massive clean-sheet projects such as the A380 and 787.

Still, while it would be unwise to rule out further delays and cost-overruns at Crossrail, it would be a pity if its missteps became another nail in the coffin of ambitious infrastructure investment. Eurotunnel's investors probably regret that France and Britain built a crossing under the English Channel; the construction costs were 80 per cent over budget. But the rest of us are glad it exists. Ditto, the Sydney Opera House, which cost 1,400 per cent more than it should have and was worth every cent.

Rather than shrinking from grand dreams, we should learn from the mistakes. Prof Flyvbjerg recommends breaking up large projects into less risky chunks. Being transparent about costs, and any overruns, is essential to preserve public trust.

The shaky-looking case for a high-speed rail link connecting London with Birmingham and Manchester has, for example, been undermined by claims it could end up costing double the already eye-watering forecast of £56 billion.

If we're to meet the expectations of the planet's 7.6 billion inhabitants while curtailing and adapting to climate change, we'll need trillions of dollars more infrastructure this century. And some of it will have to be big. Crossrail shows that with political will and ample capital, humans can do pretty extraordinary things. Sadly, though, it's usually not as quick or cheap as we might wish. BLOOMBERG

  • Chris Bryant is a Bloomberg Opinion columnist covering industrial companies. He previously worked for the Financial Times.