Let’s be courageous, part 10

(… continued from Let’s be courageous, part 9, posted on 13 Sep 2016).

Another whole subsection of the section that deals with the business case (Section Two) in the Taxpayer’s Alliance briefing paper, Rich man’s toy: The case for scrapping HS2, is used by author Harry Fairhead to comment upon the methodology that has been employed to forecast the level of passenger demand for HS2: under the subheading Forecasting demand he makes the charge that this methodology involves “inherent uncertainties” that  imply that “the BCR analysis is unconvincing”.

In order to appreciate why Mr Fairhead should think this, it is necessary to understand how the analysts working on the HS2 business case have approached their task.

The basis of the demand prediction is a projection into the future of passenger demand for the entire railway network of Great Britain, assuming a “Do Minimum” approach, i.e. implementing committed rail enhancements, but not constructing HS2. The modelling employed is a sophisticated computer algorithm dubbed “the PLANET Framework Model” (PFM), which employs, in turn, a number of separate forecasting models. The basis of the PFM method is the division of Great Britain into 235 “zones” and the prediction of travel demand, by road, rail and air, between each zone and each other (see footnote 1).

At the heart of this model lies the realisation that “it is not reasonable to expect rail demand to grow indefinitely” and that “at some point in the future, it is likely that the market would saturate” (see footnote 2); this situation is illustrated by the family of curves drawn with dotted lines in the diagram reproduced below.

Source: HS2 Ltd (see footnote 3)

Source: HS2 Ltd (see footnote 3)

The diagram illustrates how demand for long-distance rail travel, measured for example by trips per year, is expected to increase over time. The four dotted-line curves illustrate the demand characteristic that results depending upon whether the market starts a decline into saturation early, the brown curve at the bottom, or late, the green curve at the top: which of the four curves typifies the most accurate prediction for the future demand for long-distance rail travel in Great Britain is, of course, anybody’s guess.

Despite its undoubted sophistication, the PFM makes a fairly basic simplifying assumption in order to represent this tailing off of passenger demand growth, which is to characterise the demand curve by the figure shown in full black line on the diagram (see footnote 4). This figure comprises two distinct regions: an initial period of growth at a constant annual rate, indicated by the sloping line at the left-hand end; and, a subsequent period of perpetual zero growth, represented by the horizontal section at the right-hand side. The area contained under the acute angle so subtended represents the total demand for long-distance rail services over a chosen time period.

Given that the base year demand is known, this simplified demand “curve” may be defined by three parameters: the average annual increase in demand assumed for the initial growth period, the demand level at which growth is assumed to stop, and the year that growth stops (although this third parameter may be determined from the other two).

The constant annual increase in demand that has been assumed for the initial growth period for the “Do Minimum” works out at 2.2 per cent per annum (see footnote 5). The report of the Economic Affairs Committee of the House of Lords (EAC) points out that this growth rate is taken “across all the zone-to-zone pairings in the model that are over 100 miles”, and that “some zone-to-zone forecasts are higher, some are lower” (see footnote 6).

HS2 Ltd points out that the average growth rate that has been assumed “is lower than the recent trend”, which it claims “has equated to an average year-on-year growth rate over the past 18 years of 4.9%” (see footnote 5). This claim is broadly supported by a graph reproduced in the EAC report, which shows that “the number of journeys on franchised long-distance train operators increased from 77.2 million journeys per year in 2002/03 to 129 million journeys per year  in 2013/14”; this equating to “an average annual increase of 4.8 per cent” (see footnote 7).

The EAC report also refers to a claim made in evidence given by the Transport Secretary that the business case had used “a very conservative estimate about future growth” (see footnote 8). The EAC report comments that this claim “is not borne out” by a figure reproduced in the report that shows the growth rate for individual years over the same period varying between around 10.5 per cent and 1 per cent and with a distinct downward trend since the peak in 2006/07 (see footnote 9).

The EAC report also refers to an interesting alternative way of looking at this assumed growth rate, provided in evidence by Alison Munro, Managing Director – Development, HS2 Limited. Referring to the number of annual long-distance rail trips per-person that the HS2 Ltd demand forecasts implied, she commented that it “increases from 2.1 at the moment to about 2.9, so we are not talking of a massive change in the way that individual people behave” (see footnote 10).

Their Lordships appear not to have been swayed by this argument, pointing out that it represented “a 38 per cent rise in the average number of long-distance trips per person per year”, which their report characterises as “a substantial increase by any standard” (see footnote 11).

(To be continued …)


  1. An appreciation of the complexity and degree of sophistication of the PFM may be gained from a glance at the substantial publication PLANET framework model (PFM v4.3) Model description, HS2 Ltd/Department for Transport, October 2013. For each of the zone to zone travel combinations the model is capable of applying a different set of economic assumptions.
  2. See paragraphs 5.4.2 and 5.5.10 in the publication The Economic Case for HS2, HS2 Ltd/Department for Transport, October 2013.
  3. The diagram is Figure 14 in The Economic Case for HS2.
  4. HS2 Ltd apparently acknowledges this to be a shortcoming of the PFM. In paragraph 5.5.11 of The Economic Case for HS2, we are told that a “more gradual representation of the approach to market saturation is used by the Department for Transport for forecasting future aviation demand”, whereby “maturity is … introduced gradually”.
  5. See paragraph 5.2.2 in The Economic Case for HS2.
  6. See paragraphs 84 and 85 in the report The Economics of High Speed 2, House of Lords Economic Affairs Committee, 1stReport of Session 2014-15, March 2015.
  7. See paragraphs 92 and 93 and Figure 2 in The Economics of High Speed 2.
  8. See Q222 on page 258 (Evidence Session 19, Tuesday 9thDecember 2014), Oral and Written Evidence, Select Committee on Economic Affairs, The Economic Case for HS2.
  9. Figure 3 in The Economics of High Speed 2. The comment is made in paragraph 88 of that report.
  10. See Q67 on page 388 (Evidence Session 6, Tuesday 28thOctober 2014) in the Lords EAC Oral and Written Evidence.
  11. See paragraph 88 in The Economics of High Speed 2.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: