Another blast from the past that needs updating:
The interesting thing is its now emerging there were two flaws found in the process, the current one that Virgin complained about in an independent report submitted 5 days before the award (seen by channel 4), dismissed by the Dft but then found by the PWC audit, and a seperate unknown 'minor flaw not expected to affect the result' that the Dft found a week before the award and Greening commissioned PWC to investigate (reported by Conservative Home). This PWC investigation while working on another task then uncovered the flaw of interest not being included in the computerised risk model (Guardian/BBC). Its also funny to watch how the different news outlets attack the story from different angles and find different insider sources and different aspects of the story in their investigations.
The PWC report does not mention interest. It found several flaws with the process, one of which is that DfT used Real prices (the "they forgot inflation" one). A second is a more technical issue, probably wouldn't have affected the outcome, but I wouldn't call minor as it would have affected the size of the size of the risk (of all parties) and therefore any possible SLF. A third is that
the DfT failed to use their model, instead basing the number on a calculation derived from the "ready reckoner", which is not a model, but a way to get a rough estimate of the model output
once you have determined the level of risk. On top of this it appears that DfT made undocumented changes to the levels of SLF required from First & Virgin.
I don't know what the independent report contained, but the above means that the two main areas of the Virgin Legal Claim: that DfT failed to follow their own processes, and that they failed to treat bidders in the same way, have both proven to be true. This is why the SoS took the actions he did.
Pure speculation on my part, but these were so accurate that I can't help but wonder if Virgin also have a DfT source, maybe one that wasn't happy at the process.
Also we know it didnt affect the other two bids because the flawed computerised risk model was only used on the Virgin and First Group bid, a different evaluation was done as a first stage which eliminated the other two. It seems now the reason First's figures agreed with the Governments figures is because they used the original ITT documentation to recreate the model and so theirs didnt include inflation either, Virgin meanwhile in backwards engineering the dft model did include inflation in theirs. So looks like First arent at fault afterall, they just were stupid enough not to notice inflation wasnt accounted for when backwards engineering the model in the same way the Dft didnt notice.
As I said above DfT didn't use their model at all in looking at the SLFs. They used the "ready reckoner", which as I never tire of saying, is not a model. What it was though is equally flawed, as it was based on the flawed outputs from the model. All bids were subject to the SLF check and all bidders were informed of their result. It seems that two of the bidders were told that no SLF was required.
Now we come down to First. All bidders had the "ready reckoner" but again, as I never tire of pointing out, that does not allow you to calculate a value of the SLF. Indeed Laidlaw is very scathing of the DfT due to not making the SLF calculation transparent. Laidlaw also says that bidders should have been able to calculate their figures if they were to make sensible bid choices, which pretty much underlines my point (repeated again at the bottom) that it was impossible for bidders to duplicate the DfT model. Of course nothing would stop bidders attempting to second guess DfT, but
First would have to have made three other identical errors, as well as correctly estimating the DfT's risk number, in order to come up with the same value of SLF. Clearly nonsense. As I have said previously, O'Toole was just trying to back DfT when he claimed they reached the same number. Notice there's not been a peep out of First since on the issue. We can be confident that First either never had a model, or if they did, it almost certainly came up with a bigger number.
The position on Virgin is less clear, but they clearly did have a go at duplicating the DfT results. Europa, on behalf of Virgin, came up with a number for First of £600m. There are two ways they could have arrived at this that I can see.
It could have been based on the (erroneous) ready reckoner, the size of their own (erroneous) SLF, and the differences between the two bids, which included the important distinction that Virgin bid a 7% EBIT compared to First's 5% (a difference worth £20-60m a year). That would seem the most sensible approach, and it could have come up with the £600m number, but of course the number would be wrong since it would be founded on several errors. I'd think that correcting those errors would lead to an even larger number.
They could also have arrived at a number completely independently, i.e. by trying to replicate the DfT model. However with no knowledge of how the DfT model was put together there would be no chance of accurately replicating it (as per Laidlaw), and therefore no chance of duplicating the DfT result except by blind chance. You might go this way if you wanted to argue that your risk model was better than DfT's, but that wasn't what Virgin were arguing. They were saying that the DfT's own model should have shown £600m. That's why I think they will have followed a process close to what I outlined above.
). If Labour wouldn't do it in thirteen years then the Conservatives certainly aren't going to do it now. Unless Labour suffer a sudden and rather radical shift to the left I can't seem them doing it either in 2015 (or whenever they get back into power). I think we're going to end up with either a bodge job trying to hold the present system together or concessions à la Merseyrail and London Overground.
