Let’s be courageous, part 12

(… continued from Let’s be courageous, part 11, posted on 7 Oct 2016).

The analysts working on the HS2 business case set the demand cap at an annual total of 290,146 long-distance rail trips over 100 miles (see footnote 1). In the oral evidence that he gave to the House of Lords Economic Affairs Committee (EAC) Professor Peter Mackie supported the application of a demand cap to the growth prediction, but described the decision on where to set this cap as “a bit of an arbitrary assumption” (see footnote 2). Not only is it “arbitrary”, it may also be the case for the HS2 business model that the choice that has been made is largely unsupported. The published documentation  that explains the October 2013 revision of the business case justifies setting the cap at around 290,000 trips per year on the basis that it “is based on the number of trips originally predicted in the economic case originally published in February 2011” (see footnote 3), but my searches of the latter documentation have, so far, failed to yield any enlightenment of why that demand level was deemed suitable then.

The authors of the documentation that was published in October 2013 in support of the revision to the HS2 business case acknowledge that it “is difficult to predict with any certainty when long-distance rail market saturation might occur”, but contest that their analysis “does not suggest that … [the demand cap scenarios assumed] …  are implausible” (see footnote 4). This appears to be a very lukewarm endorsement, as there can be a very large distance indeed between not implausible and likely. Further, the justification employed is the increase in trips per person that Alison Munro had cited in oral evidence to the EAC that had signally failed to convince their Lordships, as I reported towards the end of part 10 (see footnote 5).

In the Forecasting demand subsection of Section Two of his briefing paper Rich man’s toy: The case for scrapping HS2, Taxpayer’s Alliance Policy Analyst Harry Fairhead claims that “it is very significant at what point the demand cap is set and reflects the uncertainty of the business case”. He supports his assertion by the results of some calculations that are reported in a paper by a researcher at the University of East Anglia, but the original source is analysis presented in the documentation that was published in October 2013 in support of the revision to the HS2 business case (see footnote 6).

For this analysis HS2 Ltd carried out sensitivity testing using a variety of scenarios, but all employing the assumption that the demand cap was 20 per cent lower than had been assumed for the standard case: this has the effect of moving the year in which demand is capped from 2036 to 2027. With this reduction applied the point-estimate benefit cost ratio (BCR) for the full network with wider economic impacts (WEI) included, drops from 2.3 to just under 1.5, moving it from “high value for money” to “low value for money”, and two-thirds of the scenarios modelled rate as low value (see footnote 7).

Conversely, a similar analysis using the assumption that the demand cap is set just 10 per cent higher, moving the year in which demand is capped from 2036 to 2040, increases the point-estimate BCR to 2.8 and 95 per cent of the scenarios modelled yield high value for money (see footnote 8). If the demand cap is increased by 39 per cent, delaying its effect until 2049, the point-estimate BCR soars to 4.5 (see footnote 9).

In view of the apparent lack of justification for the demand cap value that has been employed for the standard case, it is hard to reject the thought that the cap level may have been selected by working back from what was viewed as an “acceptable” BCR value – in the terminology employed by HS2 Ltd, this is surely not an “implausible” proposition.

In the subsection that follows Forecasting demand in Harry Fairhead’s briefing paper, which he has called Demand based upon GDP, he comments that there is a “very strong link between long-term GDP growth and the value for money of HS2”. This virtually echoes HS2 Ltd’s own conclusion, based upon similar sensitivity testing to that employed to investigate the relationship between the value for money assessment and the placing of the demand cap (see footnote 10). Mr Fairhead’s interpretation of the results of this sensitivity resting is that “if long-term GDP growth is between 1 per cent and 2.25 per cent, the most likely scenario is that it will achieve low value for money” (see footnote 11).

Using the same sensitivity testing results, Mr Fairhead is able to claim that “the chance of achieving high value for money is much higher if long-term UK GDP growth is between 2 per cent and 3 per cent per annum”. The $64,000 question is whether such comparatively high rates of growth will be sustainable over the extended period covered by the HS2 passenger demand prediction (see footnote 12).

In written evidence to the EAC, the HS2 Action Alliance rather undermines the whole premise by contesting that the link between passenger demand increases and GDP growth “no longer exists”, citing the example that demand for long-distance rail travel “has grown during the [2008-2013] recession, but stagnated when the economy came out of recession and subsequently” (see footnote 13).

(To be continued …)

Footnotes:

1. To be precise, the cap is set at the year where the annual total predicted by the demand modelling comes closest to the target number of trips; this year is 2036. In 2036 the predicted number of trips is 288,469. See paragraph 2.3.2 in the publication The Economic Case for HS2 PFM v4.3: Assumptions report, HS2 Ltd/Department for Transport, October 2013.
2. Professor Mackie is Professor Emeritus at the Institute for Transport Studies, University of Leeds. For the context of his comment see under Q4 on page 666 (Evidence Session 1, Tuesday 14^th October 2014) in the Lords EAC Oral and Written Evidence.
3. See paragraph 2.3.3 in The Economic Case for HS2 PFM v4.3: Assumptions report.
4. See paragraph 5.5.5 in the publication The Economic Case for HS2, HS2 Ltd/Department for Transport, October 2013.
5. See paragraph 5.5.6 in The Economic Case for HS2.
6. The results of this analysis are summarised in Figure 11, Figure 12 and Figure 13 in The Economic Case for HS2.
7. See paragraph 5.5.7 and Figure 13 in The Economic Case for HS2. It is also helpful to compare Figure 13 with Figure 1 in that document.
8. See paragraph 5.5.1 and Figure 11 in The Economic Case for HS2.
9. See paragraph 5.5.3 and Figure 12 in The Economic Case for HS2.
10. See paragraph 9.7 of the publication The Economic Case for HS2: Value for Money Statement, Department for Transport, January 2012. The precise words used in this document are “there is a very strong correlation between GDP growth and the value for money of the proposal and much rests on expectations of long-term economic growth”.
11. Mr Fairhead’s observation is based upon Chart 1 in The Economic Case for HS2: Value for Money Statement. This chart shows a family of histograms superimposed upon each other, with each histogram relating to a different range of long-term average GDP growth rate from 1.50-1.75% to 2.75-3.00%. The degree of saturation of the colour fill of each histogram I used to distinguish between them, with the degree of saturation reducing as the GDP increases. This chart is for the Phase 1 BCR calculation only, and it is not clear if WEI have been included.
12. That this is the case is apparent from the histograms to the right-hand side of Chart 1 in The Economic Case for HS2: Value for Money Statement.
13. See paragraphs 53 to 63 in HS2 Action Alliance—Written evidence on pages 428 to 442 in the Lords EAC Oral and Written Evidence. The comments cited may be found in paragraphs 58 and 61.