05.05.17
Railway costs through the roof?
Lord Berkeley, chairman of the Rail Freight Group, gives his thoughts on the National Audit Office’s (NAO’s) recent report into modernising Great Western Main Line (GWML) and the escalating costs of the project.
The recently published NAO report on ‘Modernising the Great Western Railway’ made depressing reading for those who want a cost-effective and growing railway. The cost of electrifying the GWML from Paddington to Swansea and Bristol has risen from £800m to probably £4.2bn, although the government tries to obfuscate these figures by reducing the scope and lately arguing that, since all the IEP trains will now have diesel motors in them, some of the original sections to be wired can be omitted with minimal effect on performance.
However, adding diesel motors comes at a price, not included in these estimates. It is understood that there was a cost of adding diesels as an optional extra on the original order; it was X million per train if ordered before September 2014, and perhaps 5X if ordered later. So even though it was obvious well before then that the wires would not be up in time for the delivery of the trains, the DfT waited until months later to change the order, paying 5X instead of X.
Why have the costs rocketed to this extent? There are many reasons: DfT changing its mind on the specification or scope, and telling Network Rail to make the change but not put it on the budget as an extra; ‘leave it until later when ministers and officials will have moved on!’. Network Rail also did not know its asset in enough detail to design it effectively, and seemed to forget that it would need several thousand permissions for work not covered by permitted development rights.
Worst of all, with a Regulatory Asset Base (AKA a limitless credit card) there was no incentive to price work effectively, or to ensure keeping to budget. The result is a serious lack of expertise in basic costing and estimating within the company. Much of this happened on Simon Kirby’s watch; he then moved to HS2 where, perhaps unsurprisingly, the same inability to estimate costs is more than evident; expert QS Michael Byng costed HS2’s design of Euston station and subsequently the whole of phase 1, and has demonstrated to HS2, officials and ministers that the cost is £48bn and not £24bn. This means that, since the “government remains committed to delivering HS2 within this funding envelope” of the £55.7bn, of which £7bn is for rolling stock, there will be no HS2 beyond Birmingham and Hansacre unless the Treasury finds some more money.
The problem is not just about underestimating, because there are many examples of where Network Rail has quoted such high costs of enhancements, perhaps for fear of pricing too low, that the project does not go ahead. I am not, of course, blaming Kirby personally; the problem goes back many years in a failure of governance both in Network Rail and the DfT, who after all have to sanction expenditure by Network Rail. The real issue is that neither Network Rail nor HS2 know how to estimate properly.
What are the possible solutions?
We have to find them since, otherwise, enhancements or new works will not happen. Ministers are clearly fed up with all this and are encouraging the industry to challenge costs, and come up with alternative structures for developing new or upgraded lines. East West Rail is a good example, but how much liberty will the new team led by Rob Brighouse be allowed in designing, getting permissions, building and operating a line separately from Network Rail? Will they be allowed to avoid the ridiculously expensive and time-consuming GRIP process beloved by those in Network Rail wedded to process rather than progress?
Why shouldn’t sponsors and financiers prepare a three-page specification and put it out to competitive tender for design, get permissions, build and operate the infrastructure? An early example of this may be March to Wisbech reopening, where such an approach might save 30% of the cost estimated by Network Rail.
Beyond all this is the need for a number of estimators who know how to do it, trained to use the RMM suite of railway cost measurement commissioned by Network Rail from Michael Byng but sadly not yet adopted by them.
Let us see a ‘can-do’ approach to change and improvement, always asking the question ‘can it be done cheaper or better or quicker’ rather than starting with the old railway response to anything new: ‘well, there are 25 reasons why this will not work or cannot be done’ and ‘it’s more than my job’s worth!’