tim's experiment

Well I hope none of you were surprised to hear that the great new White Elephant High Speed Railway to Shenzhen is running two years late already. Large projects of this kind often succumb to what is called the Planning Fallacy: over-optimistic assessments of the prospects for early and below-budget completion. Academic controversy continues over whether this is due to an innate bias in the human brain towards optimistic projections of the future, or it should more properly be blamed on unscrupulous contractors who make unrealistic promises, knowing that once the project has started it is unlikely to be cancelled however bad the news gets.

I must congratulate the MTR on contributing to the great tradition of offering limp excuses for rail delays, pioneered by the UK’s former Southern Region, which was often paralyzed by autumnal avalanches of fallen leaves. The MTR’s problem, it seems, is that it rained. What a shock that must have been. Also the tunnelers ran into some rocks under Lai Chi Kok. You can imagine the two great men in their Baker Street flat musing over this item. “See, Holmes, according to the newssheets people who were digging a hole under Lai Chi Kok found rocks there.” The great man took his pipe from his mouth. His lip curled. “Indeed, Watson, and what were they expecting to find, marshmallows?”

The great international example of this sort of catastrophe is the Sydney Opera House. Work on this started in 1958, at which time the budget was A$7million. When the building opened in 1973 it had cost A$102million. A more recent example is the building which houses the Scottish Parliament. Work started in 1997, with the projected cost at GBP40 million. When it was completed in 2004 the estimated final cost was GBP431million.  Rail projects are the specific hobby of a Danish prof called Bent Flybjerg. He been trying to foster more pessimistic planning by beating people over the head with the past statistics, like these: between 1969 and 1998 the average over-estimate of passenger numbers for major new rail projects was 105 per cent. The average (not the maximum, the average) cost over-run was 45 per cent. Interestingly the early casualties did not produce any more caution in the later estimates. the error rate remained the same. This suggests that it is probably the same now.

More recent international examples are equally ominous. Barcelona’s ninth underground line was supposed to cost 1.9 billion Euros. It came in at 8.7 million Euros. The modernisation programme for the UK’s West Coast line was originally supposed to cost between 2 and 3 billion GBP. Revised estimates effortlessly zoomed to 14.5 billion GBP, at which point the track company became insolvent. A new rescue strategy was brought in in 2003 which would, it was hoped,. keep costs down to GBP 8.3 million. These hopes have not been entirely realised; work is still in progress and the final bill is expected to surpass GBP 9 million.  In this company the Edinburgh tram system, which should be opening next month, looks a bargain. When work started in 2006 the budget was GBP 498 million. Final bill with financing costs is expected to be only GBP 1 billion. But this is deceptive. Nearly half of the originally planned network was cancelled to save money.

What can we hope for, then, from the MTR’s Great White Elephant? Well there was no competitive bidding for the contract (our two railway companies having been merged with each other) so the original number offered – $80 billion – can be considered an honest guess. Unfortunately this was considered politically unacceptable so it was massaged down to $68 billion. This is now the figure used in hand-outs and press reports, although the project was not trimmed in any way; all that happened was that some parts which could plausibly be inserted elsewhere in the public works budget – like roads – were excluded from the headline figure on the grounds that they “would have to be built anyway…”  The MTR has now produced a $4billion contingency fund, which is going to cover the latest glitches. So this is now presumably a $72 billion project. It is difficult to believe protestations that the two-year delay now expected will have no further impact on costs.

I infer two things. Firstly about suggestions from Shenzhen that the Hong Kong Government should accept some financial responsibility for disappointing ridership on the already existing line between Shenzhen and Guangzhou: these should be resisted strenuously. Two years may not be the end of it. Secondly there is the position of the responsible minister, who despite  being on the MTR professed himself totally surprised by last week’s news. If I were you, mate, I would resign now. It’s all downhill from here.