Further written evidence from HS2 Ltd
(HSR 169A)
HS2 LTD RESPONSES
TO TRANSPORT
SELECT COMMITTEE
QUESTIONS
Thank you for your letter of 18 July in which you
enclosed a number of questions. I attach our responses to these.
The Committee asked a question on the proposed 18
trains per hour service level. We are aware that this has also
been raised during the consultation and we have begun further
work in this area. The answers here reflect this work.
There were a few questions which we considered would
be best answered by the Department for Transport. The Department
will provide responses to questions 5, 6 and 16. I hope our responses
to the remaining questions are of assistance.
In addition, I also attach responses to the questions
raised by Oxera in its review of the Economic Case for HS2.
As with the questions put to us by the Committee, we agreed
with the Department for Transport that they would be best placed
to answer some of these. The Department will provide responses
to questions 1-4, 6-8, 14, 19, 20, 31, 32 and 34 from the Oxera
review.
The Committee should note that while these answers
reflect the evidence base and analysis that has underpinned our
advice to Government, they do not take into account any new evidence
presented via the recently completed public consultation, as the
analysis of the responses is still at an early stage. Clearly,
no final decisions will be taken by Ministers until that process
has been completed.
1. Without HS2 or a strategic alternative,
when will capacity on WCML be exhausted? Please give details of
projected load factors by year, section of the route and time
of day; and explain what assumptions they make about pricing,
including possible measures to smooth demand
The attached maps show HS2 Ltd's forecasts of demand
(daily trips) and load factors (proportion of passengers to seats)
on the WCML for the Intercity West Coast Train Operator only (currently
Virgin Trains) in 2008, 2021 and 2043.
The demand forecasts underpinning these maps show
that the total number of long distance rail trips over 100 miles
will grow by 95% between 2008 and 2043, or around 2% per year
on average.
On this basis, even allowing for substantial committed
capacity increases, the average all day load factor on the West
Coast TOC into and out of London is expected to grow from 56%
in 2008 to 75% in 2043. As this is an all day average it represents
a very significant level of crowding; during the peaks and on
busier routes the load factors would be much higher than 75% resulting
large numbers of people standing on many trains.
The reduction in the load factor shown between 2008
and 2021 on some sections of the WCML reflects the fact that the
forecast growth in trip numbers over this period is less than
the increase in capacity provided by the introduction of the "Very
High Frequency" timetable at the end of 2008 and the committed
lengthening of the Pendolino fleet. However, it is important to
note that actual growth on Virgin Trains between 2008 and 2011
of around 10% per year has been noticeably higher than the average
growth of 2% per annum forecast. This means that the demand level
forecast for 2021 has already been reached and the load factors
shown for that year are therefore likely to prove an under-estimate.
The forecasts provided assume that the Pendolino
fleet retains the same number of first class carriages as today
and that current policies on fares remain in place, including
an annual increase of RPI+3% between 2012 and 2015 in line with
decisions taken following the 2010 Spending Review.
While we have not ourselves undertaken any analysis
of possible pricing measures to smooth demand, a 2006-07 study
by AECOM for DfT which focused on peak flows at busy London termini
concluded that very little could be done in practice to shift
people out of the peak. Peak fares on long distance services are
already significantly higher than during the off-peak, and we
believe it unlikely that different pricing structures—that
is, further increases in fare levels at peak times—would
do much to smooth demand. And by 2043 crowding across the whole
day is expected to be so high that such pricing measures would
be largely irrelevant.
There is no simple definition of when capacity on
a route is exhausted. It can be defined as the point when additional
seating capacity via longer or more frequent trains can no longer
be reliably provided, or the point when existing trains become
unacceptably crowded. Further complexity is introduced as demand
and train frequencies vary considerably across different days
of the week and times of day.
This means there is no established definition of
when a route is full. However, as explained in their evidence
to the Committee, analysis has been conducted by Network Rail
which indicates that a) by the mid-2020s some train services on
the West Coast Main Line will be sufficiently full that—ordinarily—further
capacity enhancement intervention would be considered essential,
but b) no further viable options for enhancing capacity will be
available. The recently published West Coast Main Line RUS strongly
supports a new high speed line as the best solution to long term
capacity issues on this route.
2. Has the DfT's guidance been followed fully
with respect to the sensitivity tests performed on passenger demand
forecasts and income elasticities?
As far as possible HS2 Ltd has followed the Department's
guidance for forecasting, appraising and sensitivity testing transport
schemes. However, given the nature and scale of HS2, HS2 Ltd considered
that certain elements of the guidance relating to sensitivity
testing and choice of forecast year or demand cap are not appropriate.
In terms of testing the robustness of the rail demand
forecasts, the guidance recommends that forecasts are subject
to sensitivity tests regarding the elasticity of rail demand to
changes in income and fares. The impact of these sensitivity tests
is to change the rate of growth in rail demand which, in turn,
will affect the business case for a project or policy.
Since the range of uncertainty regarding the growth
in rail demand is greater than uncertainty around the income and
fare elasticities, HS2 Ltd chose to undertake more wide ranging
sensitivity tests than recommended by the guidance. The tests
undertaken by HS2 Ltd included varying the rate of rail demand
growth, the year at which growth is capped and the impact of lower
(indeed, non-existent) growth in air and car travel.
The results of these tests produce a much wider range
of demand outcomes than would be obtained from following the guidance.
The tests undertaken by HS2 Ltd therefore exceed the guidance
and remove the need to carry out the specific tests suggested.
The details of all of the sensitivity tests undertaken by HS2
Ltd are set out in Chapter 7 of the Economic Case for HS2 Consultation
Document.
3. What is the base case or do-minimum scenario
(in terms of more rolling stock or track capacity) on which the
answers to Q1 above are predicated? Can you confirm that the same
base case is used in calculating the BCRs of HS2 and the strategic
alternatives such as Rail Package 2?
The do minimum scenario includes all major rail schemes
to which Government was committed as of summer 2009. Among the
more important schemes this includes 106 additional Pendolino
vehicles on the WCML, IEP trains running on the ECML and Crossrail.
The Chiltern Evergreen 3 upgrade is not included, as it was not
agreed at that point.
By 2043 crowding on some services is expected to
become so severe that it causes instability in the demand model.
To increase confidence in model performance the capacity of all
long distance services on the ECML was therefore further increased
by 20% above that assumed by IEP, even though this does have the
consequence of potentially understating the benefits of HS2.
The base case used to assess the network-level strategic
alternatives published as part of the suite of consultation documents
included a similar allowance of additional capacity on the ECML.
Separately, in updating the economic case for the original "rail
packages", additional modelling instability was encountered
and a further adjustment was required. This was achieved by lengthening
all Pendolino services to 11 cars in the base case and is acknowledged
in the report on this update.
4. What assessment has been made in the business
case of the implications for HS2 of Evergreen 3 - the improved
Chiltern line service?
HS2 Ltd has not modelled the impact of Evergreen
3, and hence the exact impact on the HS2 business case is unknown.
However, it is our view that the implications of
including this scheme are likely to be small. Even with the upgrade,
journey times on the Chiltern route will still be longer than
the WCML and onward access options from Marylebone more limited
than at Euston. More importantly, the Chiltern route is only a
viable alternative for those passengers travelling between London
and Birmingham whereas HS2 is serving a much wider market.
We expect to be exploring this further as part of
our advice to Secretary of State following the consultation.
7. We were told by TfL that they were worried
about capacity at Euston to handle and disperse the passenger
flows generated by the combination of HS2 and existing routes.
They told us that Crossrail 2 would need to be in operation before
the opening of HS2 to handle demand. We have also received representations
about the disruptive impact of the Euston HS2 works on the existing
train service into Euston. Can you comment on these issues?
HS2 would increase the numbers of national rail passengers
using both the mainline and Underground stations at Euston. As
well as rebuilding the mainline station improvements to the Underground
station would also be made, including a new direct link to Euston
Square. These works would be expected to deal with crowding issues
in the Underground station.
In terms of crowding on the underground trains, our
analysis indicates that the extra number of Underground passengers
at Euston due to HS2 would be small compared with the overall
numbers already using the Underground. Whilst high quality public
transport access and dispersal at all HS2 stations would undoubtedly
be desirable to maximize the benefits of the new line, HS2 Ltd
is not in a position to comment on the value for money of particular
schemes.
Even without HS2, the number of national rail passengers
arriving or departing Euston will grow by 100% by 2043 resulting
in around 8,700 additional passengers using the Underground station
during the morning peak. The average loading of all London Underground
services travelling through Euston Underground in the three hour
AM peak period is currently 138% and is expected to increase to
185% in 2043 without HS2. Both the Northern and Victoria lines
which stop at Euston are likely to be particularly heavily crowded.
By 2043, HS2 Ltd estimates some 5,500 additional
rail passengers would use Euston Underground Station in the morning
peak as result of HS2 Phase 1—but this is equivalent to
an increase in the total number of passengers travelling on London
Underground services passing through Euston of just 2%. As a result
of HS2 Phase 1 we estimate only a small additional increase in
crowding on LUL services at Euston from 185% to 191%
Accordingly, any requirement for additional underground
capacity would be predominantly triggered by general growth, not
the effect of HS2. We believe that the same conclusion holds,
in principle, for the wider Y network which would further increase
demand at Euston but may also limit demand growth at Kings Cross
St Pancras as passengers transfer to high speed services from
the ECML and MML. We have not, however, analysed these numbers
in equivalent detail, as we have yet to finish our work on developing
the route and business case for links to Manchester and Leeds.
HS2 Ltd will be working with TfL on the impacts at Euston as the
Manchester and Leeds work is finalised, and is ready to provide
any necessary input to TfL's wider ongoing strategy for modernising
and improving Underground services should a decision be taken
to proceed with HS2.
Impact on existing train services into Euston
Our Euston station proposal was developed from the
outset to minimize disruption to users during its construction.
Euston station would be one of the most complicated
areas of HS2 to construct. It would be undertaken over a number
of years in several stages. Staging was developed on the basis
of constructing new platforms on the western side to provide initially
temporary platforms for existing "classic" services
during subsequent construction stages and then progressively rebuilding
the existing station in a number of stages. Upon completion of
each stage, new platforms would be bought into service immediately.
At the current feasibility stage of the project,
the proposals provide at least 14 platforms at all times. This
followed an examination of the current standard off-peak and morning
peak service. The latter demands the use of all 18 platforms at
times and represents the "worst case".
We concluded that the off-peak service could be maintained
with a minimum of 11 platforms with minor adjustments to operational
practice to reduce some long turnaround times. We considered a
12th platform would be useful to maintain all-day operational
flexibility for special events and the like.
The most critical period currently in the Morning
peak is 08:40-09:10, with all 18 platforms being in use for part
or all of this period. In order to operate the current volume
of trains with a minimum of 14 platforms, we concluded turnaround
times for some services would need to be reduced in line with
those achieved at other similar London terminals. This would require
detailed changes to arrival and departure times. However, by so
doing it would be possible to maintain the current quantum of
services, their frequency and general service intervals.
The London and South East Route Utilisation Strategy,
July 2011 (Chapter 4, Table 4.1), notes that Euston currently
operates less trains per hour in the peak than other London main
line termini which have a similar mix of long distance and commuter
services. This reflects extended platform occupation for some
services. For example, the Highland Sleeper arrives before the
critical peak but the empty stock remains occupying a platform
until 0926 at present.
Therefore through the period of Euston development
we concluded that the existing train service could be maintained.
Passengers would experience some level of inconvenience through
building works changing pedestrian, public transport and road
access during the multiple stages of the works, similar to other
major rebuilding projects of which the most recent and relevant
was St Pancras.
We noted also in our report that on occasions it
would be necessary to close the station entirely for short periods
in order to stage the work. This would typically be to disconnect
the track from one set of platforms and reconnect to another set,
with the accompanying alterations to the signalling and overhead
electrical equipment. Such closure would probably last several
days, and would need to be timed to coincide with periods of lower
demand such as long Bank Holidays or during the summer. The extent
and timing of such closures would be developed if the project
proceeds and be planned in conjunction with the availability of
alternative routes.
8. Do you foresee any capacity problems on
WCML north of Lichfield if HS2 opens in 2026 but the second phase
to Manchester (and Leeds) does not start to open until 2032? Can
the northern part of the line accommodate traffic from both HS2
and the classic line, including freight?
We do not foresee any significant capacity problems
on WCML north of Lichfield, subject to the completion of Network
Rail schemes as set out in the Technical Appendix to our 2009
Report.
We considered the effect of the Day One service on
the capacity of each route section of the WCML north of Lichfield
including both routes towards Manchester (via Crewe and via Macclesfield).
Our baseline was the existing capability of the routes concerned
including any relevant NR committed schemes.
We concluded that sufficient capacity would exist
to operate the proposed pattern of HS2 services without detriment
to other Train Operating Companies and Freight Operating Companies.
However we do acknowledge that detailed timetabling work for a
potential 2026 timetable would be necessary and would expect to
work with Network Rail should the project be developed.
We based our conclusion on the service pattern modelled
for Day One and the change it would represent, as shown below:
| Existing Trains per Hour from London
| Assumed Future
Trains per Hour from London
| Difference |
| London to | Pendolino
or Voyager
| Pendolino
or Voyager | HS2
| Total | |
| Crewe | 0 | 1
| 0 | 1 | +1 |
| Chester | 1 | 1
| 0 | 1 | 0 |
| Liverpool | 1 | 0
| 2 | 2 | +1 |
| Manchester via Stoke | 2 |
1 | 2 | 3 | +1
|
| Manchester via Crewe | 1 |
0 | 1 | 1 | 0
|
| Glasgow | 1 | 1
| 1 | 2 | +1 |
| | |
| TOTAL | +4
|
The above table compares the potential future off-peak services
with the existing off-peak standard hour. In peak hours we have
modelled a number of additional services, broadly in line with
the level of service frequency enhancement in peak hours that
already exists on the WCML. Further detailed work would be necessary
to produce validated timetables at a later stage of the programme.
The effect on each route section would be:
Lichfield to Colwich
This section of the WCML was modernised and four-tracked as part
of the West Coast Route Modernisation project, completed in December
2008. Sufficient capacity exists on this section to accommodate
the four additional HS2 services without detriment to other operators
or freight services.
Colwich to Cheadle Hulme via Stoke
The London—Manchester via Stoke line diverges from the WCML
at Colwich Junction. At Stone the line from Stafford converges,
bringing with it three trains per hour (two Cross Country and
one London Midland), and further north there are also local services
linking Macclesfield with Manchester.
We concluded that the addition of an additional path per hour
was achievable although detailed adjustments to the timetabling
of existing services might be necessary.
Stafford—Crewe Liverpool and Scotland
We concluded the proposed additional services could be accommodated.
From Stafford through to Crewe the existing railway is four-track.
Whilst Crewe station itself is complex due to the series of flat
junctions, there is some grade separation of the freight lines
which bypass the station. Between Crewe and Winsford the railway
is four-tracked, reducing to two-tracks going north apart from
a four-track section between Hartford and Acton Bridge, where
freight trains can be overtaken by passenger services.
North of Warrington we proposed an increase of London services
from one to two per hour. Having considered the limited capacity
available we proposed that the two services (one HS2 and one classic
rail) would be flighted closely together. This would maximise
the connectional possibilities between the two services and minimise
the demand on capacity.
9. We have been told that no high speed railway currently
operates at 18 trains per hour. To what extent is 18 tph dependent
on the deployment of new signalling technology? What are the main
risks to meeting that target? Does HS2 Ltd have a strategy for
risk mitigation? If 18 tph is not achieved, what implications
would this have for services frequencies, destinations served
and the economic case?
We calculate that an 18tph service can be offered reliably on
HS2 on a "clock face" regular interval timetable. This
is based on using the European Train Control System (ETCS), the
most up-to-date available design for the reference train, our
proposed configuration of infrastructure for stations and junctions,
and our proposed service pattern. We do not consider it requires
technology development to achieve this at 225mph and we believe
that only limited, foreseeable, development would be necessary
for the ultimate potential route maximum speed.
Capacity of current HS railways
The operational capacity of any railway depends on traffic type,
speed and stopping pattern mixes, layout of junctions and stations
and train technical performance as well as the innate capability
of the signalling control system. For example, the Japanese centralised
train control system permits trains to run safely at three minute
intervals (notionally 20tph). However due to constraints of terminal
station capacity from initial construction in 1964 and a complex
pattern of stopping, semi-stopping and non-stopping trains, including
trains timetabled to overtake each other, the practical capacity
is generally limited to 13tph.
Existing high speed lines have been designed and are operated
with a variety of national systems developed variously in the
1970s and 1980s, such as TVM in France and LZB in Germany. These
are reported to give a capacity of up to 16tph but in practice,
sharing use existing lines into terminal stations this figure
is reduced to 12 or 13 tph.
New high speed lines in Europe are now being designed and fitted
with ETCS (European Train Control System) which is now mandated
under Interoperability legislation. Examples are Malaga—Madrid—Barcelona
and Milan—Rome—Naples.
The HS2 proposal
The HS2 proposal was developed as a whole operational system to
accommodate the specific intensive frequency required, having
taken full account of experience elsewhere on factors limiting
practical capacity.
We have determined capacity on the HS2 route based on:
— the current functionality
of ETCS Level 2 train control system and the standard European
train radio system, GSM-R.
— the acceleration and braking
performance of the HS2 Reference Train which used the performance
data from Alstom for its AGV third generation high speed train
with 360kph (225mph) capability.
— all trains having the same
design performance.
— the gradient profiles of
the proposed route where there is an absence of sustained steep
gradients, which affect train braking distances, due to the topography
of the land traversed.
— avoiding capacity pinchpoints
by providing grade separation at junctions and in Euston station
throat, acceleration and deceleration tracks and high speed turnouts
to minimise capacity loss at junctions, optimum design of track
sectioning for the detection of train position and sufficient
through platforms at Old Oak Common.
— no
intermediate stations hence no stopping or overtaking movements
on the two track core section.
On the basis of the technical performance of both
the ETCS/GSM-R train control system and the Reference Train:
— the minimum headway between
successive trains operating at the maximum initial speed of 360kph
would be 116 seconds and 11.6km. The emergency stopping distance
of the reference train from 360kph would be 5.7km.
— the
worst case minimum headway between two trains was established
as when a southbound train joined the line at a junction to follow
a train passing through at full permitted speed. Following subsequent
acceleration of the second train it would be some 130 seconds,
and 13km, behind the first.
A realistic combination of though and stopping trains
at Birmingham Interchange, and through and diverging / converging
trains at the point on the "Y" network where the northwest
and Northeast limbs join, would give a technical capacity of 29
tph with existing available technology.
To convert this to a practical, timetable planning,
capacity we took into account general accepted railway practice
which allows for typical day to day service variations, and European
guidance published by the International Railway Union (UIC). On
this basis we calculated a maximum planned capacity of between
21 and 22tph.
The proposal then took into consideration the potential
variability of on-time performance of trains coming on to the
HS2 route from existing railway lines. In the first stage the
greater proportion of trains would be joining off the West Coast
Main Line at Lichfield. In the second stage a higher proportion
of train services would operate wholly on the dedicated high speed
network but a significant number would still start on existing
railway.
Having regard for the future desire for regular "clock-face"
services, with some principal destinations served on a 20 minute
cycle, we determined that we should plan HS2 service specifications
on the basis of 18tph.
Risks to delivering reliable 18 tph capacity
The principal risks of adopting the European Standard
ETCS with GSM-R as the radio carrier are of technical obsolescence.
However we considered that the continuing development of the system
in the years before it was brought into service on HS2 would more
likely enhance its capability than diminish it. Similarly, the
risk of the European Union abandoning ETCS for a less capable
system was considered remote. In any event we considered the additional
margin we had already created by confining to 18tph compensated
for this theoretical risk.
We consider other developments of train and signalling
control technology would more likely enhance the ability to deliver
high capacity reliably. For trains we expect to see improved guaranteed
braking and deployment of automatic train operation (ATO) under
driver supervision. The former already exists on some trains without
recourse to novel unproven technology to the degree necessary
to maintain the headways, and therefore capacity, described above
at 400kph.
ATO has been widely used on metros and a version
is operated on Japan's high speed lines, but there is not a specific
version for ECTS Level 2 currently. It is anticipated that others
will have created and proven such a product before 2020. We see
the main benefits of ATO as being around improved management of
key junctions to minimise potential delay and reduce energy consumption.
Implications of not achieving 18tph
Although we consider we have adopted a prudent approach
in using up to 18tph to develop the illustrative service specification
of HS2, we have considered the impact of a lower planned capacity.
We would expect to maintain the range of destinations and key
frequencies by joining trains south of the West Midlands (ie two
x 200m trains joined together) where demand forecasts indicate
no requirement to operate full 400m long trains.
10. Does the proposed land-take for HS2 provide
sufficient width for widening to 4 tracks at some future date?
Are structures being designed with future widening in mind?
No. The reasons for this were discussed in HS2 Ltd's
2009 report High Speed Rail: London to the West Midlands and Beyond.
The relevant extract is reproduced below:
"6.1.60 We have considered the possibility of
laying four tracks along the HS2 alignment, thereby increasing
its long term capacity, to 30+ train paths per hour in each direction,
rather than the maximum of 18 we assume for a two track railway.
However, for the following reasons we have not pursued this option:
— The
level of demand we have assumed—with growth continuing to
2033 and then levelling off—does not appear to support such
a substantial increase in capacity. That is not to say that the
demand may not materialise at some point after the next 25-30
years. However it does suggest that HS2 would need initially to
bear a substantial degree of additional cost and environmental
impact on the basis that the demand may materialise. We do not
believe this to be a credible position.
— Were
a second trunk into London to be justified, there are compelling
reasons to believe that its optimal alignment would not follow
the same as that of HS2. Firstly, a more easterly leg would enable
high speed rail to address a broader market, and bring with it
the possibility of further improved journey times to destinations
east of the Pennines. Secondly, we believe there is no plausible
site for the approximately 20-platform station that would be required
to serve a four track high speed railway operating at full capacity,
and nor is there a surface alignment into London that could support
a four-track HS2. Were HS2 to be four-tracked to a point somewhere
outside London, the second pair of tracks would need to enter
London entirely in tunnel and terminate at a different station
(although it is worth restating here that our analysis of the
possible sites in central London suggests there is no obvious
location for a second high speed terminal). A further reason for
preferring a second route alignment is the added resilience it
would give to a national network.
6.1.61 In short, a decision to four-track HS2 would
need to be made against the backdrop of considerable uncertainty
about future demand, and when other potentially superior options
exist should the demand materialise. For these reasons we have
recommended as part of our proposals for HS2 that it remain only
a two-track railway. If demand materialises, a second leg could
be built from the East Midlands to London."
11. Pierre Messulam, SNCF Director, told us
that although passenger forecasts may prove broadly correct in
the long run, the timing of costs and income can be problematic
and make a significant difference to project viability. What sensitivity
analyses have been performed on HS2 cost and income profiles?
Costs and revenues can affect project viability in
a number of ways. HS2 Ltd has undertaken a number of sensitivity
tests around the effect on the business case if demand growth
were to be higher or lower, or capped at a different level, some
of which are described in response to question 2 above.
The nature of cost and revenue risks also differs
under different models for funding and operating high speed rail.
In particular, revenue risk would be a key issue for a private
sector concessionaire under a PPP arrangement such as has been
used for recent TGV projects in France.
Ernst and Young provided some early views to us on
financing issues for our 2009 Report, High Speed Rail: London
the West Midlands and Beyond and, for example an illustrative
profile of government funding is shown in Figure 5.1b of the Report
(p203), but the work was not detailed enough to identify specific
risks around timing.
However, no decision has been taken on the preferred
financing model for HS2. More work would be undertaken in this
area if the decision is taken to take the project to the next
stage.
12. How do the costs of a single route via
Heathrow compare with the costs of the HS2 route plus a spur to
Heathrow?
We have reported on the estimated base construction
costs for a route via Heathrow with a station at Iver and the
HS2 route plus a spur to Heathrow in our report to Government
published in March 2010 and also in our Options for Connecting
to the Heathrow Area Supplementary Report published later in September
2010. Our worst case assumption for a spur, assuming a Terminal
5 station on-airport in a box below-ground was that the estimated
base construction costs would be c£200 million less than
for a route through a Heathrow Hub 4km north of the airport. As
noted below we calculated the journey time penalty to a through
train of stopping on a route via Heathrow Hub as 7 minutes.
The cost of the spur option would be incurred, under
the government's presently stated intention, with the second phase
of HS2 to the North East and North West, when direct airport services
would be scheduled.
We continue to work with BAA to investigate the possibility
of creating a ground level station at Terminal 5. Should this
be the case, the cost of the spur would reduce and therefore the
cost differential between it and the more expensive through route
would increase.
13. We have received a detailed submission
saying that the Heathrow Hub / Iver Common site assessed by HS2
Ltd is not the same as the Heathrow Hub station proposed by Heathrow
Hub Ltd, and that the conclusions drawn are therefore misleading.
Can you please comment?
Briefly, we believe the site we assessed is at the
same location as Heathrow Hub Ltd's, although the assumed nature
of the station and services are very different.
During preparation of our report in 2009 we met the
promoters of the Heathrow Hub on a number of occasions to discuss
their proposals and they made a written submission to HS2 Ltd
to outline their proposed scheme.
We believe that the location we assessed, adjacent
the Great Western Main Line and the M25 on flood plain land occupied
currently by the Iver Sewage Treatment Facility, is the same as
that proposed for the Heathrow Hub. We also assumed that it would
be linked to the airport by an extended people mover.
However, there were a number of differences in approach:
— Our remit was to provide
an interchange for HS2 with Heathrow Airport and Crossrail. Accordingly,
we assessed only the benefits that such an HS2 interchange would
bring. In contrast, the Heathrow Hub proposal was developed as
a wider interchange with conventional rail and coach services.
— The Heathrow Hub proposal
included the shared use of the high-speed rail corridor into London,
such as by "Javelin" services to Kent. These were not
compatible with the capacity that we considered would be needed
for long distance high speed services on HS2, consuming around
half the potential capacity of the HS2 route.
— The Heathrow Hub proposals
envisaged a new 30 million passenger airport terminal integrated
with the hub.
— In
addition, our station design and approach routes differed slightly—we
did not consider that a surface level station was viable given
the necessary tunnelled approach either side.
However, setting aside these differences, we did
not progress this station location because:
— If
it was served by a through route, as the Hub promoters proposed,
passengers on HS2 travelling to and from London—who comprised
the great majority of travellers—would incur a significant
journey time penalty.
— For
air passengers it was not as attractive as options closer to or
on-airport.
— For
HS2 passengers going to and from London it was not as attractive
as a Crossrail interchange at Old Oak Common and we could not
envisage these passengers transferring to Crossrail at this location,
thereby increasing demand at Euston.
14. If a route were to be provided via Heathrow,
what is your latest estimate of the time penalty that would be
incurred by stopping trains at Heathrow? Has this figure been
used in your economic analysis?
The economic analysis assumes that the journey time
penalty of going via Heathrow would be an extra 7 minutes including
a station stop.
15. The business case analysis assumes fare
levels similar to those on the existing WCML. If premium fares
were charged on HS2 (as occurs on HS1), what would be the impacts
on passenger numbers, the business case and the level of much
public subsidy. How has the business case taken account of the
prospect of fares competition from Chiltern Line and WCML?
The demand model used by HS2 Ltd makes use of data
on average fares between any two places across all routes and
services. It is therefore unable to analyse any competitive effects
of different prices across different routes or operators.
Nevertheless, for the work published in March 2010
HS2 Ltd undertook a high level study on the impact of charging
a premium fare for HS2 trains. This found that the impacts of
premium fares are many and complex, however it is certainly not
the case that a single percentage premium is applicable or desirable
across all markets and routes. Instead there would need to be
careful management of revenue strategies—similar to those
already seen on long distance services—to maximise use of
capacity. The initial conclusions from this work were:
— The
scope for premium fares is complex, with the impact varying considerably
across different markets and routes.
— The
availability of non-premium service alternatives would make a
big difference to the scope to generate additional revenues through
premium fares.
The economic case does not take account of fares
competition across different routes or operators and we have not
sought to undertake this analysis. Amongst other things, the impact
of competition will be dependent on the regulatory framework under
which HS2 would operate
More work would be needed in all these areas if the
Government decides to proceed with the scheme following the public
consultation.
17. What analysis have you made of the use
of HS2 by income groups and how does this compare with current
rail use?
HS2 Ltd has not specifically undertaken this kind
of analysis. However, as the economic case assumes that HS2 trains
are priced at the same level as classic rail services then it
would be expected that the income distribution of HS2 will be
broadly similar to the existing income distribution of users of
long distance services classic rail.
OXERA: REVIEW FOR
THE CASE
FOR A
HIGH SPEED
RAIL PROGRAMME
HS2 LTD RESPONSES
5. Has further work been completed to improve
the robustness of the [economic] case for the Y network?
The current consultation is focused on the strategic
case for the overall Y network, but only a detailed proposal for
the initial London—West Midlands phase. Therefore, a strategic
assessment of the potential economic case for such a network was
appropriate.
Work is currently ongoing to develop a detailed route
proposal for extending the network to Leeds and Manchester, which
would form the basis for subsequent public consultation. The work
involves developing a detailed line of route that will allow a
much more accurate assessment of the costs, benefits and environmental
impacts of the Y than was presented in the recently completed
consultation process. HS2 Ltd will provide advice to Government
on the refined Y route later this year.
9. How dependent is the business case on the
standard forecasting framework? Have alternative (especially non-PDFH)
rail demand forecasting frameworks been tested?
HS2 Ltd has not directly tested the impact of different
demand forecasting frameworks.
Instead, the robustness of the economic case to different
levels of demand has been tested by adjusting the overall rates
and level of demand growth. These included varying the rate of
rail growth, the year at which growth is capped and the impact
of lower (indeed, non-existent) growth in air and car travel.
Further explanation of these tests is provided in the response
provided to question 2 of the Transport Committee's questions
to HS2 Ltd.
10. How was the level of the demand cap determined?
What evidence is there to support it being set at the level selected?
There is no evidence to suggest the observed trend
of strong growth in long distance rail travel is slowing. Nevertheless
HS2 Ltd considered that it would be prudent to cap demand at some
point in the future and that it would be reasonable that this
cap should be defined at a particular level of demand, rather
than at a particular point in time.
With no evidence available to suggest whether or
when demand might saturate HS2 Ltd undertook its own analysis
of existing trip rates by examining how many long distance trips
people currently make across different income quintiles of the
population. This showed that a demand cap that results in a growth
of 61% more long distance rail trips (over 100 miles) per person
per year in the absence of HS2 seems reasonable. Such a cap is
equivalent to about 0.5 extra long distance rail trips per person
per year and represents an average rate of trip making (1.4 rail
trips per person per year) across the whole population that is
lower than the rate already seen in today's population containing
the top 20% of incomes (2.0 rail trips per person per year).
Taking account of population growth we forecast the
total number of long distance rail trips (over 100
miles) to increase by 95% by 2043, representing an increase of
1.9% a year on average. This compares
to a growth of around 5% a year over the past 15 years.
This demand cap is also consistent with the forecasts
originally published in March 2010, when demand was capped in
2033. Since then amended assumptions about higher fares and lower
economic growth mean that this level of demand is not now forecast
to be reached until 2043.
The Department for Transport's current guidance is
that demand should be capped in 2026. This assumption is not based
on the belief demand growth will cease in 2026; it is just a convenient
simplifying assumption that allows the Department to easily compare
different conventional rail schemes on a consistent basis. However,
HS2 Ltd consider that a project of HS2's size, timescale, and
strategic impact necessitates thinking much further ahead than
2026 and that the Department's normal guidance is therefore not
appropriate. The Department for Transport and HMT agreed with
and fully supported this change in approach.
11. Have other scenarios of higher or lower
fare increases been tested?
Yes. Scenarios involving both higher and lower fare
increases have been tested and this is reported in the Demand
and Appraisal Report published on the HS2 Ltd website.[272]
The results of these sensitivity tests are set out
in the table below. It should be noted that while the overall
benefit:cost ratio for the initial London-West Midlands phase
of HS2 is 2.0 including Wider Economic Impacts (WEIs), the sensitivity
tests carried out by HS2 Ltd were calculated as variations on
the BCR for the proposed line without WEIs and therefore
show a "central case" BCR of 1.6.
| Test | Socio-economic
benefits
| Revenue | BCR
(excl. WEIs)
|
| Central case | 16,544 | 13,660
| 1.60 |
| High rail fares (cap 2064) | 13,521
| 19,598 | 3.09 |
| Low rail fares (cap 2035) | 18,131
| 11,449 | 1.45 |
| High rail fares (cap 2043) | 11,156
| 11,566 | 0.90 |
As a rule, higher fares will reduce demand and lower fares will
increase it. However, the effect on the benefits and revenues
in the business case is more complex. In particular an assumption
must be made about how fares impact the cap on demand.
It is consistent with the approach we have taken to the cap on
demand, explained in Question 10, to assume that the cap occurs
at the same level of demand as the central case reached
in 2043. In this case higher fares reduce the rate of demand growth
with the demand cap now reached in 2064 rather than 2043. That
reduces benefits (fewer people save time before 2064) but increases
revenues (after 2064, higher fares result in higher revenues).
Similarly, lower rail fares increase demand and the cap level
is reached earlier, in 2035. That increases benefits (more people
save time before 2043) but reduces revenues (after 2043, revenues
are lower). As the table shows, in these circumstances the net
impacts are that higher fares result in a higher BCR and lower
fares a lower BCR. Implicitly these scenarios assume that a change
in fares levels will change the rate of demand growth but not
the overall level of demand that is ultimately reached.
In contrast, if it is assumed that the cap on demand always occurs
at 2043 (the same time as in the central case) irrespective
of the fare or level of demand, then the result of higher fares
is to reduce the level of demand in 2043 and all later years compared
to the central case. This has the effect of significantly reducing
both benefits (fewer people save time) and revenues (fewer people
pay fares) after 2043. The net impact is to reduce the BCR. Whilst
a sensitivity test has not been carried out on this basis for
lower fares it is likely that they would have the opposite effect—eg
increasing benefits and revenues and raising the BCR. Implicitly
these scenarios assume that changes in fare levels will both change
the rate of demand growth and the overall level of demand growth
achieved.
12. Has further work on premium fares for high speed services
been conducted since 2010? How would premium fares affect expected
revenues? In particular, is there a role for demand management
to "lock in" the benefits of the High Speed Rail programme?
Please refer to the response provided to question 15 of the Transport
Committee's questions to HS2 Ltd.
13. Has the benchmarking suggested areas where costs could
be reduced through efficiency savings?
Yes. Our benchmarking work has specifically resulted in reductions
to assumed tunnelling rates, which are included in our most recent
cost estimate. Our work also became a key input into the Infrastructure
UK cost study looking at the costs of UK civil construction projects.
The IUK's conclusions were reported at the end of 2010 and were
that there was potential to reduce the cost of building such projects
by at least 15%. Were this potential to be realised, we could
see a saving in HS2 scheme costs of nearly £5 billion for
the "Y" and an improvement in BCR from 2.6 to 3.4.
15. How have asset renewal rates been derived?
We considered the individual asset types and their typical lifespans.
As a result, we made assumptions regarding which assets would
require either partial or full renewal over the 60 year appraisal
period. Where renewal was required we applied the appropriate
percentage of the estimated initial construction costs for these
asset types at the required renewal frequency in our business
case model. The renewals assumptions are shown below (see overleaf):
| Permanent Way | Full replacement by end of 30 and 60 years, each taking four years 25% spend each year
|
| Switches and Crossings | Full replacement by end of 30 and 60 years, each taking four years 25% spend each year
|
| OHLE | Renew 50% each by year 15, 30, 45 and 60, each taking two years
|
| Power supply | Renew 50% each by year 15, 30, 45 and 60, each taking two years
|
| Signalling | Renew 50% each by year 15, 30, 45 and 60, each taking two years
|
| Communications | Renew 50% each by year 15, 30, 45 and 60, each taking two years
|
| Stations | 40% renewal by year 40 taking four years
|
| Earthworks | No renewal in evalution period
|
| Retaining Walls | No renewal in evalution period
|
| Structures | No renewal in evalution period
|
| Tunnel | No renewal in evalution period
|
| Depot / stabling | 50% renewal by year 30 and 60 taking three years
|
| Trains | Renewal by year 35 spread in line with initial expenditure phasing
|
16. What progress has been made in improving the robustness
of the cost estimates for the Y network?
We are undertaking work to develop route proposals for the Y during
2011. This enables a better understanding of the assumed scope
of work which is fundamental to further refining the Y network
cost estimates. We have also engaged Davis Langdon to provide
specialist independent cost support during 2011. Their work includes
review of unit rates, quantitative risk analysis and improvements
to the granularity of cost estimation enabling more detailed challenge
than was previously possible.
17. How have the cost savings on the conventional network
been estimated?
HS2 Ltd has calculated the net change in train kilometres for
each type of rolling stock running on the classic rail network
as a result of introducing HS2. This change includes both a reduction
of longer distance services and an increase in shorter distance
services. Information on the cost per train kilometre of different
stock types has been obtained from the Department for Transport's
Network Modelling Framework, and this is then applied to the calculated
change in kilometreage. This is more fully described in Annex
2 of the document "A Summary of Changes to the HS2 Economic
Case" published on the HS2 Ltd website.[273]
In line with DfT guidance, an optimism
bias uplift of 41% is then applied to the net change in operating
cost as measured across the whole of rail network (thus including
both HS2 operating costs and classic line savings).
18. Has there been analysis equivalent to
that of Graham and Melo (2010) for the agglomeration benefits
from additional commuter capacity? If so, does it provide indications
of the robustness of the estimate of £3 billion agglomeration
benefits?
HS2 Ltd has followed the Department for Transport's
standard guidance for calculating agglomeration benefits. Other
than the Graham and Melo work, HS2 Ltd has not undertaken any
additional work to test the robustness of this approach for the
calculation of agglomeration benefits.
The Graham and Melo work was undertaken to specifically
investigate whether there were any additional benefits on top
of those already captured within the standard guidance due to
the unusual characteristics of high speed rail.
21. Would it be possible to reduce the estimated
range of potential carbon emissions?
The range of potential carbon emissions estimated
in the report is wide because there are a large number of uncertain
factors outside of the scope of the HS2 project. Whilst it would
be possible in theory to reduce the range, this would rely upon
the uncertainty being reduced, either through better information
or by more specifically defined assumptions. For example, Government's
firmer policies on future GRID energy mix will help to refine
the estimates.
22. Will the estimated carbon values in the
AoS be updated to reflect this change?
The carbon valuations in the AoS will be updated
to reflect the most current guidance and reported to Government
by the end of the year.
23. Is it correct that there may be a net
increase in carbon emissions because there is no reduction in
the number of flights and additional HS2 services?
The assessment of carbon impacts included in the
Appraisal of Sustainability for HS2 includes a wide range of uncertainties,
including around the impact on aviation. This indicates that if
no reduction in flights were to occur as a result of HS2, then
depending on the carbon intensity of electricity generation used
to power HS2 trains, there would be an increase in carbon emissions.
It is important to note, however, that even in the worst case
scenario, where the carbon intensity of the grid remains as it
is today and no modal shift from either aviation or road is achieved,
the average increase in annual emissions would still be less than
1% of current annual transport emissions. In addition, HS2 Ltd's
forecasts indicate that the full Y-shaped network could see as
many as six million trips a year transfer from aviation, meaning
that it is credible to imagine that some air services would be
discontinued.
It has also been argued that an increase in carbon
emissions could occur where domestic flights are cancelled as
a result of passengers shifting from flights to HS2, and freed
up slots are then refilled by new flights of a more polluting
nature (eg domestic flights being replaced by international flights).
However, in this scenario, caps on emissions would apply due to
the introduction of aviation into the EU Emissions Trading System.
As a result, there would not be a net increase in carbon emissions
across the System as a whole.
24. If changes in the cost of carbon beyond
2050 are considered, would this significantly affect the estimates?
The effect that the cost valuation schedule introduced
by DECC in 2010, which extended the end year from 2050 to 2100,
would have on the AoS carbon cost valuation is not simple. For
a given level of emissions, the cost of carbon would increase.
This means the range estimated in February 2011 would widen.
However the cost of carbon has real impacts on the
economy and on choices made by the travelling public because of
the impact of the EU Emissions Trading Scheme. This means a higher
carbon price is likely to increase electricity costs, air fares
and other prices. This may change transport decisions which in
turn could change the impact of HS2 on carbon emissions.
25. Will these assessments be carried out
when more data becomes available in order to have a better indication
of the impacts, and would these assessments be likely to have
a significant impact on the BCR of the scheme?
The cost valuations will be reassessed using the
most current guidance, to report back to Government this year.
Going forward, if Government decides to progress scheme planning,
some of the uncertainties in the carbon assessment will be reduced,
for example through greater clarity in Government energy policy;
level of detail in the design proposals; and, commitments to different
construction or operation strategies. Carbon assessment would
continue to be a theme through all future design and environmental
assessment. We would expect revisions to be within the range considered
and even in the worst carbon case, it is unlikely to have more
than a small impact on the BCR.
26. Have estimates been made of the landscape
impacts of a new high speed line, and would these be likely to
have a significant impact on the relevant BCRs? Are such assessments
planned?
The Appraisal of Sustainability has considered landscape
impacts in line with DfT guidance; and this was a consideration
throughout the route selection process.
In 2009-10, HS2 Ltd undertook an indicative assessment
of the potential impact as the route stood at that stage. This
valued the impacts at up to £4.5 billion, but was a brief
assessment to consider potential scale of impacts and has not
been updated to reflect the latest design and mitigation of environmental
impacts. The Department is also reviewing, in the light of recent
evidence, how analysis of landscape impacts can be best undertaken
to provide robust measures of such effects.
27. Has there been consideration of the types
of user, in terms of socio-economic status, who will benefit from
travel on HS2?
Please refer to the response provided to question
17 of the Transport Committee's questions to HS2 Ltd.
28. Are there expected to be significant distributional
effects between socio-economic groups as a result of the construction
of the HS2 line?
The Appraisal of Sustainability provided a high level
assessment of the potential impacts on the most deprived people
in the community, as identified by the Government. Impacts on
equality would be addressed at a greater level of detail in future
stages of scheme planning, should Government decide to proceed
with the scheme.
29. What estimates have been made of the adverse
economic impacts on areas not served by the new high speed line?
In the relatively few places where HS2 might adversely
affect the transport provision of an area (either through poorer
train service provision or through higher road congestion), the
impact is captured within the overall net benefits presented within
the economic case. As explained in question 30 below, no analysis
has been undertaken of how such effects are distributed.
30. What is the relative size of the economic
impacts on cities expected to be served by the high speed network?
What proportion of these economic impacts is abstracted from other
regions not served by the high speed network?
HS2 Ltd has not undertaken analysis of how different
regions or cities will be either positively or negatively affected
by HS2. In part how different cities and regions may or may not
benefit is related to a wide range of issues and local policies.
HS2 Ltd has examined how benefits are distributed
on the basis of where trips originate, and on this basis well
over 50% of benefits fall outside London and the South East. Around
one third of benefits accrue to trips starting north of Birmingham
with the North West the biggest beneficiary.
We are aware that a number of regional cities have
carried out their own analyses of the potential economic impacts
of new high speed rail lines and that the key outputs have been
included in their evidence to the Committee.
33. How would substantial long-term oil price
rises or falls have an impact on demand for rail? Would the impact
be greater than those in the tested fuel duty scenarios?
The central case for HS2 uses assumptions on car
running costs (itself a combination of assumptions around fuel
costs, duty and VAT, vehicle efficiency, and non fuel costs) provided
by DfT's standard WebTAG guidance. This assumes that fuel costs
themselves increase by 19% between 2008 and 2043.
HS2 Ltd have run a sensitivity test in which fuel
duty is assumed to increase by 50%. This results in a greater
demand for rail services and significantly improves the business
case for HS2. (The BCR for London—West Midlands excluding
wider economic impacts moves from 1.6 to 2.4)
30 August 2011
272 http://www.hs2.org.uk/assets/x/78304 Back
273
http://www.hs2.org.uk/assets/x/77820 Back
| 
