APPENDIX 4
Memorandum submitted by The Railway Consultancy
Ltd
1. BACKGROUND
1.1 This note is in response to your request
for submissions to your inquiry into the above, but I would be
happy to provide further clarification if required.
1.2 I write as Managing Director of the
Railway Consultancy, a specialist railway planning consultancy.
My personal experience stretches back 20 years, to the completion
of a PhD at Newcastle University in railway fares policy, carried
out in collaboration with what is now the Regional Rail sector
of British Rail. Subsequent to that I spent eight years in strategic
and business planning roles at London Underground. I have run
the Railway Consultancy for the last ten years, during which time
we have carried out a range of commercial projects for TOCs and
others, notably including a study for the Rail Passengers' Committee
into the economic viability of a National Railcard.
2. INTRODUCTION
2.1 There are three criteria which should
underpin any rail fares policy. Rail fares should be related to:
(i) the cost of providing the service;
(ii) the willingness of users to pay; and
(iii) the complexity of the system.
Any consideration of the current (or any alternative)
system should bear these in mind.
3. THE COMPARATIVE
COST OF
RAIL FARES
3.1 The real level of rail fares in any
country can only reflect
(i) the level of Government support, relative
to the level of train service provided;
(ii) the efficiency of the operation.
Although British Government support to the railways
ballooned a couple of years ago, in line with West Coast Main
Line and Railtrack troubles, it has historically been at a much
lower level (expressed as a percentage of Government expenditure,
rail support has typically been only half that of our European
neighbours). That is a political decision, taken in a country
which has generally adopted the view that users (rather than taxpayers)
should pay for public services. That logic is questionable if
(as with the railways) those services indirectly benefit the entire
community, and where the railways have been prevented from taking
advantage of indirect benefits (eg British Rail was forced to
divest itself of property-related income).
3.2 Over the years, a considerable amount
of research into fares has been undertaken by British Rail and
its successors, and condensed into the industry "bible"
The Passenger Demand Forecasting Handbook (PDFH). Individual train
operators use that as the basis for commercial decisions, tempered
by the regulatory environment, which provides caps for increases
in baskets of fares. However, as with any industry, mistakes do
occur. A particular misunderstanding has occurred amongst some
train operators, who have failed to understand the ways in which
passengers respond to fares increases. The rate at which passenger
demand falls off when fares are increased is termed the fares
elasticity ef, and a critical value of this occurs at ef =-1:
at smaller values, increases in fares lead to increases in revenue,
but at larger values, increases in fares lead to decreases in
revenue, as the higher fares deter more people from travelling.
In fact, many long-distance fares elasticities in Britain are
around the -1 mark, which is profit-maximising, and therefore
has considerable economic sense.
3.3 However, some train operators have not
understood (or have chosen not to understand) a key corollary
of the theory, namely that elasticities effectively increase over
time. In the short term, whilst some passengers may immediately
be deterred from travelling by train, switching mode immediately,
others may pay up; in the longer-term, however, people start making
different decisions about home and job location, and more passengers
desert the railways. For a TOC with a short franchise, large fares
increases may still gain net revenue, with the negative long-term
effects accruing to a subsequent franchisee. Some of the complaints
(correctly) made against TOCs reflect this process, and it is
not clear how the industry should respond to such erroneous decisions.
3.4 A similar issue arises regarding the
image of the railway. If one TOC sets very expensive fares, the
public's overall view of the railway is incorrectly biased against
it. Unfortunately, it is difficult to see how a cap could be applied,
since we know that passengers are both affected by the total amount
of the fare (which, quite reasonably, varies with distance) and
by a consideration of the fare per mile (which, by definition,
does not).
3.5 At a perceptual level, of course, elasticities
reflect what passengers see as "good value for money".
Services with higher quality (eg shorter journey times, more comfortable
trains) can sustain higher fares—but that does lead to complexity
in the fare system. Whilst the Government remains of the "user
pays" opinion, one has to accept that train operators will
vary fares according to service quality, sacrificing the potential
benefits of fares being directly linked to the distance travelled.
3.6 Passenger perception is also highly
coloured by the fact that the main competitor to the railways,
the private car, is too cheap, as many of its externalities (eg
pollution, accidents, policing) are not fully borne by its users.
Moreover, there is a good case for saying that most transport
services are under-priced, especially given their dependence upon
limited natural resources. Providing a realistic basis for the
pricing of other modes, by using the tax system more effectively
to encourage use of the more efficient modes, is key; road pricing
and carbon taxes are two possible ways of doing this.
3.7 In conclusion, there is considerable
evidence that some specific fares in Britain either are or have
been too expensive, but it is not clear how to deal with this.
More generally, though, rail fares are probably at an economically-sensible
level, and apparent problems actually result from the under-pricing
of (in particular) car and air modes.
4. FARES STRUCTURE
4.1 In an industry with substantial quantities
of fixed costs, the careful management of capacity is essential.
Despite the flexible working adopted in some offices, there are
huge peaks in the demand for rail travel (demand in peak hours
often being as much as 10x those in interpeak hours). Any incentives
to encourage passengers not to travel in peak periods are therefore
essential.
4.2 As noted above, variations in fares
charged should also take into account passengers' willingnesses
to pay, which are largely a function of journey purpose. As the
peak demand is in the morning, when the vast majority of trips
are for "earning" purposes (eg commuting, business),
it is economically sensible to charge higher fares at those times.
Unfortunately, travel patterns in Britain are becoming more complicated,
and many commuting and business trips now occur at other times;
however, whilst those trips are not directly contributing to peak
capacity problems, offering an incentive to travel offpeak is
probably justified.
4.3 How one segments the market, however,
to reflect the three key criteria, is very important, and a number
of methods are used. Time of day, a requirement to book in advance,
and class of travel, are all relevant. However, it should be noted
that railcards are one method which have been found to be justified
on financial, not social, grounds.
4.4 Research I carried out as long ago as
1986 showed that substantial differences between peak and offpeak
fares are essential, if significant numbers of passengers are
to be encouraged to travel offpeak. My research showed that the
Ordinary fares at that time needed to be around 40% more expensive
than day returns/Savers. With passengers' Values Of Time having
risen since, that figure is likely to be higher.
4.5 There then needs to be a balance between
the level of sophistication of the fares offered, and the complexity
of that system. At one extreme, one could set a fare for each
individual passenger, based on their willingness to pay for the
journey, but such a policy clearly fails the "simplicity"
test. My judgment would be that there is room for about five different
products, ranging from the most expensive "do anything"
ticket through those offering choice outwith peak periods to those
restricted to specific trains. Clearly, for local markets, the
latter types of ticket are not applicable.
4.6 The position we actually have in Britain,
however, has degenerated through the disaggregation of the rail
industry. Whilst individual TOCs can aim to carry out appropriate
fares policies, the current system has a number of completely
unnecessary disadvantages. These include:
(i) the use of differing names for the same
product; and
(ii) the use of differing (eg time-of-travel)
conditions applied to equivalent products.
4.7 Even with a range of ticket types, however,
it would be possible to offer a range of types which were related
more obviously to each other than at present eg fare type a =
fare type b less 10%. From a wider industry perspective, this
would be highly beneficial, but ATOC has struggled in such areas.
4.8 Elsewhere, there are a number of ways
in which fares can be simplified. Southern's standardisation of
local fares within South London is to be applauded, whilst elsewhere
many long-distance tickets cost the same to a whole area eg London—Teesside.
This is to be encouraged, as it reduces the need for passengers
to re-book en route.
4.9 Whilst there may be some rationale for
offering cheaper tickets through sales channels which are cheaper
to operate (eg the internet), ticket sales costs are not a dominant
cost within the industry. Instead, many such products are offered
because of the internal accounting rules of the industry, which
can make it financially-worthwhile to avoid the all-operator revenue
allocation system, which is not a good reason.
4.10 In summary, there is good theoretical
justification for a limited range of ticket types, but there is
considerable scope for simplifying the current range, probably
without any loss of revenue.
5. THE AVAILABILITY
OF ADVANCE
PURCHASE TICKETS
5.1 The requirement to book ahead has been
used for many years in a number of industries (both transport
and otherwise) as a market-segmenting device. There is indeed
a clear potential benefit to the operator of receiving income
earlier, and it can be helpful for planning purposes, especially
in dealing with peaks through adjusting quotas. However, booking
ahead policies effectively assume that those able to book in advance
are likely to be less willing to pay, and should therefore be
offered a discount—but this does not necessarily follow.
For instance, there are many business trips (eg annual meetings,
conferences) which are known about significantly in advance, and
where the railways are therefore offering cheap facilities to
those who are prepared to pay more. This is not good economics.
5.2 Some TOCs have started offering really
cheap tickets available only a few days in advance. Even when
limited to specific trains, this seems unlikely to maximise revenue,
unless of course the specific trains concerned are known to be
very lightly-used, when there may be a marginal marketing benefit.
6. NETWORK RAIL
TIMETABLE IMPACTS
6.1 The provision of "book-ahead"
tickets clearly only works if it is physically possible to purchase
tickets at the relative time in advance. That requires operators
knowing when their trains are going to run. We share the view
of many, that problems at Railtrack were inexcusably the root
cause of difficulties with Advance Purchase tickets during 2004.
It is difficult to see what TOCs could have done about this (eg
in making cheap fares available without their attached seat reservations),
given that in some cases TOCs were not being told until only a
few weeks beforehand at what time their trains were running—or
whether their trains were running at all. Inclusion or strengthening
of penalty payments in the contracts between Network Rail and
TOCs, if NR fails to provide timetable data in sufficient time,
might be one way of incentivising NR to perform.
6.2 As it is, the railway is still not providing
ticket purchasing as far in advance as some of its competitors—and
this can cause problems in the holiday market. For instance, booking
a return flight to Inverness to coincide with a holiday booking
in the North of Scotland can be undertaken much before the railway
is able to do so. The industry needs to improve in this area if
it really is to compete properly in the market. The rigidity in
engineering possession strategy that this might imply can, over
time, be overcome through investment in such facilities as bi-directional
signalling which would reduce the need for line closures. Such
a system would bring Britain's railways more into line with those
in Europe, which are able to offer high-quality rail services
on all days of the week. Such an ability is going to be essential
if Britain's railways genuinely aspire to provide a service to
meet the needs of the British people, who wish to travel on every
day of the week.
30 September 2005
|
