This is from a letter by Bruce Weston of HS2 Action Alliance to the Department for Transport. Reproduced with permission
A Department for Transport spokesman told the Telegraph:
“Our modelling has been calculated using conservative assumptions and a well established approach to demand forecasting that is recognised across the transport industry, and that conforms to the highest standards of evidence.” Telegraph 3 December 2011
Your Department’s modelling is far from conforming to the ‘highest standards of evidence’ as the modelling is based on out of date forecasting factors. The version of Passenger Demand Forecasting Handbook used (4.1) is importantly out of date in regard to the income elasticity distance growth factors. These cause DfT’s model to overestimate long distance rail demand compared to the current version (5.0). The treatment of the distance term, ie its exclusion, has been confirmed by a major study conducted by Oxera and Arup. Your Department has refused to release the report from this study in response to an FOI request, but this particular result was presented to a transport economists group in February 2011.
In fact when Philip Hammond came to office in 2010 he inherited a draft Tag Unit (3.15.4d) that was to reduce these distance terms as a temporary measure  on the basis of an explicit recognition that the version 4.1 values are wrong. A full revision awaited implementing the outcome of the Oxera Arup work. The recommended maximum values in the draft guidance are lower than those used for HS2, and would have resulted in forecasting lower demand.
The draft guidance was due for adoption in April 2010. To date the draft has not been confirmed as guidance, nor has the version 5.0 approach been adopted, nor has the Oxera Arup research been published nor acted upon.
This falls well short of my concept of conforming to the ‘highest standards of evidence’.
 Unit 184.108.40.206d says:
‘4.2.9. We agree that the PDFH 4.1 recommendations produced unfeasibly large elasticities over long distances. However, in light of the ongoing research described above we are reluctant to suggest
changes to our demand forecasting methodology that may be superseded within a matter of months. As
way of compromise we recommend that PDFH 4.1 elasticities continue to be used but that a maximum
limit is placed on the elasticity value, in line with most practitioners use. In the absence of further
evidence we suggest limits of 2.5 (to London) and 1.5 (from London). Should promoters wish to use the
PDFH 5.0 elasticities then they may do so as a sensitivity test to this core analysis.’