Supplementary memorandum submitted by
the Association of Train Operating Companies
RESPONSE TO SUPPLEMENTARY QUESTIONS RAISED
BY THE TRANSPORT SELECT COMMITTEE IN RELATION TO ITS INQUIRY INTO
TRAIN FARES AND TICKETING
1. INTRODUCTION
We have set out below our responses to the Transport
Select Committee's supplementary questions in relation to its
inquiry into Train Fares and Ticketing. Many of the Select Committee's
questions were substantive in nature and we have taken the opportunity
in the time available to set out the fullest possible response
in each case. We have answered questions in the same sequence
as they were posed to us in the letter from the Committee dated
1 December 2005.
This response from ATOC is complementary to
the individual responses that are being submitted by GNER and
Virgin Trains.
As a precursor to answering the Committee's
questions, we believe it would be useful to highlight the work
undertaken by the Passenger Demand Forecasting Scheme, about which
the Committee asked questions during the hearing on 30 November
2005. As we stated at the hearing, the Scheme is largely funded
by train companies but also includes within its membership the
Department for Transport, Network Rail, the Office of the Rail
Regulator and Transport for London. We have attached, as Appendix
A, a list of research projects into price elasticities either
reviewed by the Scheme or commissioned by it over the last 10
years. It demonstrates the amount of work undertaken in this area
and shows that the currently recommended price elasticities were
reviewed by the Institute of Transport Studies at Leeds University
as recently as 2005.
As we will be quoting price elasticities later
in our response, we also believe that it may be useful to set
out a layman's guide to how these values work. Elasticity measure
the effect of price changes on the volume of demand and total
revenue. If a fare price elasticity is below -1.0 (for instance
-0.3) then total revenue will be increased if prices are raised.
In other words, the reduction in demand that will result from
the price increase will be more than offset by the higher level
of fares charged. Conversely, if the elasticity is more than -1.0
(for instance -1.5) then revenue will be increased if prices are
reduced. The value of -1.0 is the point on the demand curve at
which revenue is maximised and at which either price increases
or decreases would result in a reduction in total revenue.
2. ELASTICITY
Question A. Regular passengers such as commuters
may not desert the railways right away if ticket prices are too
high, but it will affect their long-term choices of where to live
and work. Explain why the current high ticket prices won't result
in regular passengers fleeing from the railways in the long term?
As Season Tickets are tightly regulated it is,
of course, a matter for the Government as to what the appropriate
price level should be. However, we would point out at the outset
that, even at current price levels, commuting markets are heavily
subsidised and that, as we will demonstrate below the experience
of the last 10 years does not suggest that commuters are deserting
the rail network. Given the fundamental nature of the question
posed by the Committee we have taken the opportunity to answer
the question in appropriate depth.
The commuting market, which is the focus of
this question, very largely uses Season Tickets. We assume that
the reference in the Committee's question to high ticket prices
relates to our original Memorandum to the Committee, where we
highlighted that Season Tickets prices, as measured by the average
pence per kilometre paid, are relatively high in Britain when
compared to general rail travel rates across Europe. We calculated
that the average pence per kilometre for Season Tickets in Britain
is around 8.6 pence whereas the average pence per kilometre across
all European railways is between 6p and 7p. We believe that two
factors underly this:
(i) The relatively high cost of providing
services to commuters. The commuting market is characterised by
very significant demand peaks on weekday mornings and evenings,
which drive capacity both in terms of track and signalling infrastructure
and rolling stock. Outside these peak periods demand is significantly
lower and there is, in effect, an excess provision of capacity.
The resulting unattractive economics of this market have historically
led to Season Tickets being set at relatively high price levels
compared to the broad average of rail fares across Europe, although
as already noted, high levels of subsidy are still required to
support commuting markets.
(ii) The average prices paid by commuters
are also high because of the relatively short distance of most
commuting journeys (we estimate that the average journey length
of Season Ticket holders is 26 km compared to 48 km for the non
Season Ticket market). This will result in a higher than average
pence per kilometre because, as we highlighted in our original
Memorandum to the Committee, there is an inverse relationship
between distance and pence per kilometre (the longer the journey,
the lower the pence per kilometre). We believe that this distance
related effect exists on most European railways.
However, it is important to note that the relatively
high pence per kilometre for Season Tickets users is a product
of history rather than any real price increases imposed since
privatisation. Indeed due to the regulatory formula introduced
at privatisation, which was based on Season Ticket price increases
being capped each year at 1% below inflation (as measured by the
Retail Prices Index), the real cost of fares paid by customers
using Season Tickets has fallen by 8% in real terms over the last
nine years, as the graph below illustrates:
This is also reflected in the increase in the
number of commuters using Season Tickets over the same time period.
Volume has grown by 41% since 1996 as the graph overleaf illustrates:
Thus trends over the last nine years do not
indicate long term demographic changes with consequential reductions
in rail demand of the kind suggested by the Committee's questions.
In fact, conversely they suggest that at current price levels,
people are choosing to join the commuting market, possibly reflecting
related residency and employment decisions. The decline in real
price levels will also have contributed to the significant increase
in demand experienced.
Given this, the Committee may find it useful
to understand what current evidence suggests would be the effect
of future real price increases or decreases to Season Tickets.
The current Demand Forecasting Scheme recommended elasticity values
for the commuting market range between -0.6 and -1.0 (almost all
values are below -1.0). These values suggest that overall revenue
would increase if prices were raised although volume would fall.
In other words, the market is relatively inelastic.
It is clear from all available evidence that
a reduction in current price levels would lead to a decrease in
total revenue, an increase in commuting volumes (and overcrowding
on some routes) and a concomitant need to invest in additional
capacity. The overall effect would be to require an increase in
subsidy levels.
Real increases in current price levels would
reduce volume both in the short and long term (although this has
to be considered in the context of background changes to housing
costs, employment, etc, all of which have significant impacts
on commuting demand). However, it is our commercial judgment that
the current RPI+1 formula is unlikely to cause major shifts in
employment and residency patterns of the kind suggested by the
Committee in the short to medium term (5-10 years). Real increases
will increase total revenue and reduce the level of subsidy required.
Question B. The cost of rail has increased
relative to other modes of transport over the past two decades.
Will this not serve to drive passengers away from rail in the
long term?
Since the Committee hearing, we have undertaken
further work to analyse the relative cost of rail and other modes
of transport over the last 18 years (the longest period for which
we can find consistent data). The results are summarised in the
table overleaf:
Table 3
CHANGE IN TRAVEL COSTS 1995-2005
Type of Travel Cost
| Change in Real Terms
1995-2005
| Change in Real Terms
1987-2005
|
Standard Class rail fares | +3%
| — |
| All rail fares (including First Class) |
+6% | +24% |
| Local Bus fares | +15% |
— |
| Bus and Coach fares | +15% |
+32% |
| Petrol | +25% | +31%
|
| Vehicle tax and insurance | +12%
| +44% |
| Vehicle maintenance | +25% |
+43% |
| Purchase cost of motor vehicles | -37%
| -43% |
| Total motoring costs | -7% |
-5% |
| Other travel costs | +9% |
+3% |
| |
|
Source: SRA's National Rail Trends 2004, Office of National
Statistics, Transport Statistics Bulletin (Department for Transport)
September 2005. "Other travel costs" are a combination
of all other modes including air, ferry and cycling.
The analysis demonstrates that over the last 10 years the
cost of all other modes of transport has risen compared to rail
with the exception of motoring. However, even here, all elements
of motoring cost have increased with the exception of the purchase
price of vehicles, which has fallen very significantly and outweighs
the effect of the other cost increases. However, the marginal
cost of motoring, which is most likely to be considered by consumers
when making modal choices for specific journeys, has clearly risen.
A similar pattern is true for the period since 1987, although
the cost of rail has increased compared to "other travel
costs" over this period.
In the 10 years since privatisation, rail has become relatively
less expensive than all other modes except motoring, where costs
have been very significantly affected by the real decrease in
the cost of motor vehicles. The increase in rail volumes over
the last 10 years (38% growth in journeys and 40% growth in passenger
kilometres) does not suggest that passengers have been driven
away from the rail network. Indeed, conversely it suggests that
rail may have benefited by the changes in relative price levels
identified.
We stated during the Committee hearing on 30 November that
we had undertaken further work on the average price actually paid
by rail users since 1995-96. We have analysed the amount of revenue
raised by train operators in each of the 10 years since privatisation
and compared this to the number of passenger kilometres travelled
each year. This allows us to derive pence per kilometre measure
which reflects the actual cost of travel for passengers. In order
to provide a comparison with a period immediately pre-privatisation,
we have also analysed, on the same basis, revenue and passenger
kilometres between 1986-87 and 1995-96. The results of this analysis
are shown in the graph below:
The above table illustrates that the actual cost of rail
travel to passengers has fallen slightly by around 1% in real
terms since privatisation. By contrast, in the nine years preceding
privatisation, the average cost of rail travel for passengers
rose by roundly 11% in real terms.
The average cost of rail travel has fallen (by 1% in real
terms) by comparison with rail prices which have risen (+6% in
real terms) due to two factors:
(i) the average distance by customers has increased (journeys
have increased by 38% since 1995 but passenger kilometres have
increased by 40%) bringing into play the previously observed relationship
between distance and pence per kilometre; and
(ii) there has been switching between higher priced rail
fares and lower priced rail fares, partly reflecting that customers
have a greater range of fare options to choose from and partly
reflecting the price increases applied to full fare standard and
First Class tickets.
Whilst, in the absence of comparative competitive data, we
cannot draw firm conclusions from this additional analysis, it
does highlight the remarkable stability in the real cost of fares
paid by rail passengers since privatisation.
3. COMPLEXITY AND
TRANSPARENCY OF
FARES
Question C. What are the benefits of moving away from
a traditional price-per-mile system towards so-called dynamic
yield management where the price of a ticket depends on a whole
range of factors, such as the time of travel and how far in advance
the ticket was booked?
(i) Who benefits from these new systems?
The railway industry has not used a traditional price per
mile system for some considerable time. British Rail moved away
from this approach in 1983 when it first introduced its range
of Saver and Supersaver fares. These were supplemented in the
early 1990s by advanced purchase fares and the trend towards market
pricing has been continued and extended by train companies since
privatisation.
There are three main advantages that come from a more sophisticated
market pricing approach (supported on longer distance routes by
dynamic yield management) as compared to a simple mileage based
tariff:
(i) prices can be varied to reflect the related competitive
strength of rail in specific point to point markets;
(ii) prices can be varied by market segment to reflect
willingness to pay; and
(iii) demand can be managed so as to minimise overcrowding
at peak periods and encourage utilisation of spare off-peak capacity.
The latter two points can perhaps best be understood in the
context of the three principle markets that rail serves: business
travel; leisure travel; and commuting travel.
The business travel market is relatively price inelastic
(current Demand Forecasting Scheme suggested values are -0.5 to
-0.8) and is prepared to pay a relatively high price in return
for the ability to turn up and go, flexibility in regard to travel
arrangements and refunds and, in the case of first class, added
value services such as more spacious carriages and lounges at
stations. This is the primary market that full fare first class
fares and standard single and open returns are targeted at.
The leisure travel market is highly price elastic (current
Demand Forecasting Scheme suggested values are between -0.9 and
-1.5, with most values over -1.0) and price is a key determinant
in the consumer decision to travel. It is worth highlighting that
leisure rail travel, as well as having strong direct competition
from express coaches, low cost air carriers and private cars,
also has strong indirect competition from other leisure goods
and services. The majority of the market is discretionary travel
(such as visiting friends or shopping) and at the margin this
can be substituted for by any other leisure good or service. It
is in this market that Saver and other discounted walk-up and
advance purchase fares are targeted.
The pre-booking requirements and time of travel restrictions
associated with these fares are to reduce the risk of them being
bought by business travellers and to manage demand so as to fill
spare, off-peak capacity and reduce overcrowding at peak periods.
On longer distance routes, in particular, train companies actively
manage sales of these fares using the new National Reservation
Service (NRS) and yield management systems. These systems allow
demand to be managed much more precisely and overall yields per
seat to be increased.
The commuting market has already been discussed in some detail.
Season ticket prices, the primary product in this market, are,
in effect, controlled by the Government, reflecting rail's relatively
strong position in the market.
The overall effect of this approach is to increase both revenue
and volume. In addition, using yield management systems, demand
can be managed and overcrowding reduced or eliminated. There are
three principle beneficiaries from this: rail customers; the tax
payer; and train companies.
Customers benefit from there being a wider range of fares
available and, in particular, very cheap fares for travelling
in the off-peak. Customers also benefit from the fact that increased
revenue allows train companies and the Government to invest more
in the rail network. The tax payer benefits because increased
revenue means that a lower level of subsidy is required to support
the rail network. Train companies benefit because it provides
the opportunity to use their commercial skills to grow revenue
and, within the profit capping and sharing constraints now built
into most franchises, increase profits for their shareholders.
Market pricing and yield management of the kind described
fits well with the current industry model, which allows train
companies to build revenue growth into their franchise bids and
agreements. This provides a strong incentive for them to grow
revenue and volume, creating a virtuous circle that benefits all
stakeholders, as illustrated above. Franchise agreements, with
their associated profit sharing and capping arrangements and,
where appropriate in economic terms, fares regulation are the
mechanisms which allow the Government to ensure that there is
an equitable balance of the resulting benefits between customers,
taxpayers and train companies.
Question D. How do you know whether rail users and non-users
are aware of the full range of fares and ticket types available?
Train companies promote their fares widely through advertising,
leaflets, direct mail and, increasingly, email. The last method
of communication is especially effective at letting known users
and potential users know of promotional offers.
In addition, the industry provides a multitude of ways in
which customers can find out about fares:
— through the circa 1,400 staffed stations across
the network, which are obliged through regulation to provide information
impartially and accurately on the full range of fares offered
by all train companies;
— through National Rail Enquiries, which provides
impartial information on fares either through its 24 hour, seven
day a week telephone enquiry service or the National Rail website;
— through any of the website or telephone booking
services, provided by the major longer distance operators (such
as GNER, Virgin Trains, Midland Mainline or First Great Western),
which again through regulation, are obliged to provide impartial
and accurate information; and
— through ATOC-licenced rail retailers such as the
Trainline (also an impartial retailer) and 550 travel agents.
It is also worth noting that ATOC, on behalf of the DfT,
mystery shop all stations annually to ensure that they are selling
accurately and impartially. As this is a mandatory, regulatory
activity, results are reported to and vetted by the DfT. A threshold
of 96.5% accuracy/impartiality is set and train companies that
fail to meet this target are obliged to prepare and implement
action plans. Repeated failure to meet the threshold ultimately
results in penalties being imposed on the train company concerned.
In 2003 and 2004, the average results achieved by train companies
were 98.2% and 98.8% respectively. The results for 2005, currently
being calculated look like being similar. In 2005, mystery shopping
has been extended for the first time, to internet and telephone
retailers. NRES is also subject to regulated mystery shopping
to measure the accuracy of the information, which it provides.
Whilst we believe that train companies already provide many
opportunities for customers to find out information about fares,
we also accept, as made clear at the Select Committee hearing,
that more can be done. Two specific current initiatives are:
— simplification of the range of advance purchase
fares by some longer distance train companies, with the introduction
of a single advanced purchase fare type, available up to the day
before departure; and
— improvements to web retail sites to make the presentation
of information clearer and to give customers more ways of selecting
the best fare for their needs.
Question E. What is the rationale for ticket prices to
be cheaper via the internet than other media?
Currently, rail fares sold through the internet are not generally
cheaper than tickets sold through other channels, although internet
retailers do sometimes offer promotional discounts to encourage
sales.
However, ATOC estimates that information provision and ticket
retailing costs the rail industry between £400 million and
£500 million per year. Train operators would like to reduce
this "cost of sale" and see the internet as one possible
method of doing this.
Whilst initial experience with internet retailing has not
produced the economies expected, this has partly been caused by
very significant start up costs and the complexity of the rail
network compared to other modes of transport (having to provide
fares between 2,500 stations for instance, not counting London
Underground and other destinations).
Train companies believe, nevertheless, that significant opportunities
through internet retailing exist and have invested in developing
the national Ticketing on Departure (ToD) network to support internet
and telesales retailing and are now moving forward with plans
to introduce low cost electronic ticketing (Stagecoach's Megatrain
is an early example of this).
Ultimately the key to successful internet retailing will
be to reduce transaction costs, partly through initiatives such
as e-ticketing but also by increasing volumes and realising economies
of scale. In the short term the rationale for offering cheaper
fares through the internet, therefore, is to drive volume and
achieve economies of scale. In the longer term, if the cost of
sale is reduced, lower fares via the internet would reflect the
fact that this is a lower cost distribution channel for train
companies. We would like to see changes to the current regulatory
framework that would allow train companies to exploit the internet
more effectively.
It is worth noting that there are other advantages associated
with internet retailing:
— part of the market now expects to buy online and
if this facility was not offered, some sales would be lost and
revenue reduced; and
— web sales allow, with the customer's permission,
detailed data on customers to be collected, facilitating communication
of future fares' offers and promotions.
4. OPEN TICKETS
Question F. To what extent do people who "turn up
and go" represent a "captive market", that is people
who have little or no choice but to travel by train using open
tickets?
Turn up and go customers are likely to be less flexible in
their travel plans but it is incorrect to say that those customers
are forced to buy Open tickets. Walk-up Saver fares and Cheap
Day Returns are available on all longer and shorter distance routes
respectively at off peak times. On some routes, walk-up Travelcards,
SuperSavers, Network Awaybreaks, Cheap Day Singles and other route
specific walk-up fares are available for off-peak travel. Railcard
holders are also able to obtain reductions on both Open and discounted
walk-up fares.
A breakdown of walk-up sales for each of the major fare types
in 2004-05 has been summarised overleaf showing that around 70%
of "turn up and go" customers benefit from some kind
of discounted fare:
Fare Type | Revenue and Volume 2004-05 Revenue
(£m)Volume
(m journeys)
| Percentage of
Total Volume |
Open and Day Single/Return
| 994 | 171 | 30%
|
| Open and Day Single/Return—Railcard Discount 74
| 17 | 3% |
| Saver/Supersaver/Network Away break | 816
| 77 | 14% |
| Cheap Day Single/Return/Travelcards | 639
| 295 | 53% |
| Total | 2,522 | 560
| 100% |
| |
| |
Source: Rail industry systems (Lennon). Note that for shorter
distance journeys, the full price Day Single/Return is offered
instead of the Open Single/Return. The data in this table includes
both these products.
It is also worth noting that most longer distance train companies
now offer discounted advance purchase single fares. For many journeys,
therefore, customers that have completely inflexible outward travel
arrangements and need to travel during a peak period may be able
to purchase a standard single for their outward journey and a
discounted advance purchase single for their return journey.
Finally, we should re-emphasise that Virgin Trains, GNER
and Midland Mainline have all simplified their advance purchase
fares and now allow these fares to be bought up to the day before
travel, providing the greatest possible opportunity for customers
to buy in advance.
5. SAVER AND
ADVANCE-PURCHASE
FARES
Question G. Will Saver tickets not suffer the same fate
as the SuperSaver and effectively disappear if they are de-regulated?
As we have already highlighted, train companies maximise
revenue by offering a range of fares targeted at different market
segments. There is always likely to be an element of the longer
distance leisure market that wishes to turn up and travel.
Train companies will wish to cater for this market and thus
discounted walk-up fares will continue to be offered even if Savers
are deregulated. However, train companies wish to be able to have
the freedom to integrate these fares fully into their fare's structures,
allowing their yield management systems to be used to as efficiently
as possible.
It is this context that we have argued the case for deregulation
of Saver fares. It is also worth emphasising that in the very
competitive markets that longer distance train companies operate
in, rail's relatively low market shares do not provide any economic
justification for regulation.
Question H. The allocation of advance purchase tickets
in unclear. How can passengers make informed choices if the details
of quotas are not publicly available?
Yield management systems operate on the basis that the number
of seats available at any particular price level is constantly
varied before departure, in order to maximise the number of seats
sold, whilst ensuring that enough seats are always available for
higher yield, walk-up passengers and customers purchasing more
flexible, higher priced tickets.
In such an environment there may be 100 seats available at
a particular price level on day one, 50 available on day five
and 150 available on day 10. In these circumstances it is impossible
to state the absolute number of seats available at a particular
price point on a specific train service, as this will vary over
time. Availability on a specific day could be provided but given
the dynamic nature of seat availability, as described above, this
could end up being highly misleading.
Conventional airlines and low cost airlines have been using
yield management systems for many years and do not show availabilities
on their internet sites or other sales channels. Nevertheless,
consumers understand the general principle that the earlier you
book the better chance you have of a very cheap fare. The enormous
success of the low cost carriers illustrates how powerful yield
management can be in increasing volumes, load factors and revenue
yields.
6. RAILCARDS
Question I. Would a National Railcard scheme not be the
best way to expand the market for rail travel in the UK?
We do not believe that a National Railcard scheme would be
the best way of expanding the market for rail in the UK commercially
and would very probably require a higher level of subsidy to support
the industry.
Train companies currently offer a range of Railcards targeted
at specific market segments, which research has indicated to be
more price elastic than the market in general. The Railcards are
designed to generate additional, off peak leisure travel, in particular
with a low risk of use for business travel. There are currently
around 2.2 million Railcards in use (Young Persons, Senior, Family,
Disabled, HM Forces and Network) and they account for around £400
million of revenue per year. ATOC estimates that around £60
million of this revenue is generated solely because of the existence
of the Railcards. Railcards are well supported by train companies
with around £2.5 million per year spent on managing and promoting
them.
Although targeted Railcards have been very successful, we
believe that a universally available National Railcard is likely
to lead to reduction in industry revenue. This is because a National
Railcard will provide discounts to all segments of the market,
including those that are relatively price inelastic.
Overall, we believe that a National Railcard is likely to
lead to a decrease in industry revenue with a concomitant increase
in the level of subsidy required.
We understand that the DfT, and the SRA before them, have
been considering this issue and that the SRA undertook market
research. Although we have not seen the detailed results of this
research, we understand that it generally supports our belief
that a National Railcard would be loss making.
7. EFFECT OF
TIMETABLING ON
THE AVAILABILITY
OF CHEAPER
FARES
Question J. Last year there were significant problems
for passengers in making cheap advance bookings for the Christmas
period. How do you think advance bookings are progressing for
Christmas this year?
Question K. Is the process of Network Rail uploading
timetables and engineering works data to the national database
at T-12 now functioning satisfactorily?
As they are so closely related, we will answer questions
7(J) and 7(K) together. There has been very significant progress
since last year in improving the availability of timetable information
to meet the T-12 and T-9 deadlines. Prior to this year's peak
Christmas and New Year period, all reserving train companies were
able to offer reservations, as required, at T-9 except Virgin
Trains and Hull Trains. Virgin Trains were able to offer reservations
at T-7 and Hull Trains at T-5. All reserving train operators now
have reservations open for the Christmas and New Year period.
Currently all reserving train operators are open for T-9
bookings after Christmas except Virgin West Coast (T-6), Midland
Mainline (T-8) and Hull Trains (T-6).
Train companies are generally pleased with the progress made
by Network Rail in ensuring that timetable data is available to
meet these required deadlines. However, it is important that all
parties continue to remain focused on consolidating the improvements
made to date and making further progress towards compliance with
the T-9 and T-12 deadlines, although short term engineering work
is always likely to make full compliance on all routes very difficult
to achieve. Nevertheless, we are confident that all parties are
now fully committed and that improvement will continue to be made.
8. BOOKING SYSTEMS
Question L. How widespread is it for train operating companies
to offer special assistance for business travellers helping them
to identify of the best fares and making reservations?
As we have emphasised previously, all rail customers have
access to a wide range of channels for obtaining information about
rail fares and purchasing them.
A number of train companies operate specialist business travel
units to provide specialist services to corporate customers. These
are, in effect, in house travel agencies that make bookings and
provide tickets to these customers. Tickets are generally delivered
to meet the customer's specific needs and in addition account
management services are provided. These include the provision
of regular statements that give a breakdown of employee travel,
central invoicing of the customer and the provision of credit
facilities. Some train companies charge corporate customers fees
for the provision of these services.
These services are valued by corporate customers principally
because they provide additional management information, allow
greater control over travel by employees and facilitate ticket
dispatch and collection. Whilst corporate customers will be advised
on fares, as part of the service provided, this is not the primary
purpose of these units.
Very similar services are offered by travel agents who specialise
in the corporate travel market. Rail sales through these travel
agents amount to roundly £200 million per year.
| 
