Get documents about "Creating a viable
Document Sample
Creating a viable railway for Britain –
what has gone wrong and how to fix it
Sir Christopher Foster and Chris Castles
Summary
Changes are needed in the railway industry to restore control over costs and to improve
performance. The rail privatisation inherited by the Government from the Conservatives was
flawed in crucial respects. Its design and implementation had been rushed, regulatory roles
and relationships were confused, some contractual arrangements were badly designed and
incentives misaligned. Nevertheless, rail traffic was growing rapidly, performance was
improving and investment was increasing. The accident at Hatfield precipitated a loss of
control over industry costs, largely driven by concerns over safety and risk averse behaviour
by those working in the industry following the political and media reaction to major
accidents.
To restore the industry to viability we propose few changes. We believe most structural
changes that are currently being proposed, will bring no benefit to passengers. They would
inevitably cause disruption, will be costly to implement and delay the actions by the industry
on what really matters to restoring and then improve the industry's service to passengers.
However, there are a few essential changes needed to regain control of costs and improve
performance. One is that the Strategic Rail Authority (SRA) should take the central
leadership role in the industry and the Office of Rail Regulation (ORR) should step back
from its current direct decision-taking role in specifying rail infrastructure outputs and
expenditure. Instead, the SRA should negotiate directly with Network Rail for the contracts
to supply the outputs it wishes it to provide and pay fees disaggregated by line and including
performance incentives for these outputs. These infrastructure outputs would be planned by
the SRA with Network Rail and would be complementary to the requirements of the train
operating company (TOC) franchises that the SRA is also responsible for. The TOCs would
continue to have their direct contractual relationship with Network Rail, concentrating on
day-to-day operating performance, but they would no longer be responsible for paying the
fixed element in access charges. A considerable amount of work is also needed to re-align the
incentives structure in the industry to ensure that the performance regimes are consistent with
one another and that all parties are motivated to common objectives.
For these arrangements to work, the industry needs a coherent policy framework for which
the Government must takes responsibility. The Government must provide clarity over critical
policy issues, including, the balance between road and rail, the financial limits on subsidy
provided to the railways and, within the railways, the balance of funding between subsidy
and fare box revenue, as well as the price/subsidy policy needed to match supply and
demand, the distribution of this subsidy and the size of the network and extent of rail services
compatible with the resources available This clarity is essential to enable the SRA to
develop a rational investment strategy for the railways and to specify the train operating
franchise requirements it needs. The role of the SRA is to provide the analysis of policy
options and to advise government on their implications, while the role of the Government is
1
to set the policy agenda for the SRA and to make the high level policy decisions to enable it
to carry out its tasks.
The Office of Rail Regulation would have an important role in helping to make the SRA
accountable for its decisions and efficiency. It would stand outside the operating
relationships in the industry enabling it to take an independent view of any conflicts arising.
It would act as a mediating influence, and, in carefully defined circumstances, as an appeals
body for dispute brought by industry parties, but with its decisions bound by the funding
constraints on the industry. It could also use its expertise to carry out periodic reviews of the
performance of the SRA.
Railway infrastructure costs have increased alarmingly and are now some four times higher
than before privatisation, an increase probably unparalleled in any other industry. Ultimately,
unless the railways can bring costs down to something more like historic levels, serious
questions must be asked about the scale of public resources going into the railways. The
railways are now absorbing the same level of public resources as the national roads
programme and yet the roads carry 85% of the traffic, while the railways carry about 6%.
Even taking into account relief of road congestion and environmental factors the returns from
investing in road are far greater than increased spending on the railways. The railways must
therefore restore control of costs and the proposals in this paper will provide the policy,
regulatory and incentive arrangements to enable this.
Contents
Pages
Introduction 1
The privatisation of British Rail 2– 3
Implementing the restructuring and privatisation 3-4
The lost vision of a commercially driven railway 4–8
Optimistic franchise bidding assumptions
Escalating railway infrastructure costs
Other contractual flaws
Labour’s response
No strategic leadership and confused regulatory 8 - 11
arrangements
Responding to risks 11 – 12
2
What is to be done? 12 – 21
A new role for the SRA
Development of a strategy for the railway
Regional devolution
Changing roles and relationships
New role for the Office of Rail Regulation
The incentive framework
Major Risks 21 – 22
Conclusions 22 - 23
Creating a viable railway for Britain - what has gone wrong and
how to fix it.
Sir Christopher Foster and Chris Castles1
Introduction
It is widely acknowledged that Britain‟s railways are in a mess. Train service performance
remains below the level when the Government took power in 1997 and, since the accident at
Hatfield in 2000, industry costs have escalated alarmingly so that the Government‟s strategy
of substantial further growth in the use of rail transport by 2010 now seems unachievable.
There is also general dissatisfaction with the structure of the industry and the way it operates
in practice.
The Government has therefore initiated a review of the industry. This provides the
opportunity to stand back and think carefully about what needs to be done to restore Britain‟s
railway to a viable future. It gives the opportunity for a range of views to be considered in the
debate. But it is important that the Government itself takes the lead and does not rely only on
the views of those working close to the industry. A broader perspective is required to make
the changes needed for the industry to be able to serve its customers effectively and have a
viable future.
Any solutions must be based on an analysis of the causes of the industry‟s current problems
and yet there does not appear to be a consensus on precisely what has gone wrong and why.
In our view, the fundamental problem lies with the way policy decisions are made and
translated through the regulatory and incentive arrangements in the industry. There were
1
Sir Christopher Foster has been advising on transport policy since heading the Transport Policy Unit at the
Ministry of Transport in the 1960‟s. He was a partner at PricewaterhouseCoopers and was adviser to John
MacGregor during the privatisation of British Railways, he was a non-executive Board member of Railtrack
until 2000 and was Chairman of the RAC Foundation form 1998 to 2003.
Chris Castles is a transport economist with over 30 years experience of transport policy, strategy and business
planning. He led the Transport Policy and Strategy group at PricewaterhouseCoopers for 15 years where he
worked extensively in the railway and other transport industries. He now works as an independent consultant.
(christophercastles@yahoo.co.uk, ++44207 4821596).
3
flaws in the original arrangements inherited by the present Government from the
Conservatives. Since 1997, the Government‟s actions and various initiatives have generally
undermined the incentive structure in the industry and failed to address the problems it
inherited. After over 10 years of continuous change, it is important that any future changes
in the industry are carefully thought through to ensure that they are effective. The industry is
in its current parlous state largely because it has suffered from a long record of poor quality
decisions taken too hurriedly by the governments of both political parties. Therefore we
begin by reviewing some of the key flaws in the decisions that have been taken over the past
12 years and then make some proposals for change in the future. In our view, only a
relatively few, but important changes are needed and those we propose will allow the
operating organisations to continue to get to grips with costs and performance with the
minimum of disruption. These proposals are not a complete solution but they are
fundamental to achieving the detailed changes needed to restore the structure of incentives in
the industry to enable it to operate on a commercial basis, with government support, under a
stable and coherent policy framework.
The privatisation of British Rail
British Rail was the last nationalised industry to be privatised by the Conservative
Government, largely because it was recognised to be a difficult task. Like other previously
privatised network industries, British Rail was a large, complex, integrated, geographically
dispersed, organisation with monopoly power in some markets. But, unlike the other
privatised industries, it was also loss-making. It was therefore dependent on government
subsidy (as well as some internal cross-subsidy) to retain those services that were not
commercially viable but were deemed to be socially necessary. Therefore it was bound to be
difficult for the Government to distance itself from decisions in the industry when it was
paying for their consequences. There was a high level of public dissatisfaction with the
performance of the railways and this frequently rebounded onto the Government. But any
suggestion of changes in the scope of the services or size of the network had always been
highly politically sensitive so that railway routes served had become ossified and railway
services had failed to adapt to changing market needs.
Between 1988 and 1992 there was a long debate over how BR might be privatised. The main
options that emerged were either to privatise it as a single unit; or to break it into separate
companies, either based on regions or its existing market „sector‟ businesses (InterCity,
Network SouthEast, Provincial services and Freight); or to separate an infrastructure
company from the operation of train services. All of these break-up options presented
practical difficulties, particularly with respect to separating the profitable from the
unprofitable elements of the business, targeting subsidy, deciding priorities for investment
and creating incentives for efficient operations
As the debate continued, one theme began to dominate - the need for competition within the
industry. The Government‟s earlier experience of trying to introduce competition into
telecoms and gas after privatising BT and British Gas as vertically integrated national
monopolies, led it to believe that industry restructuring was a necessary lever for improving
efficiency and customer service. On these grounds, the „track authority‟ option appeared best
able to provide for the possibility of competition by separating the management of the natural
monopoly infrastructure supply from the operation of commercial train services. Advisers at
the time warned that on-rail competition would be difficult to introduce into the railways in
Britain, both because most of the services included unprofitable elements which were
4
therefore commercially unattractive, and because the practicalities of interdependent railway
operations on a crowded network would render impracticable mechanisms, such as auctions
for „slots‟ in the timetable, to enable competition for resources and markets. It was also
feared that the fragmentation of the industry required by the restructuring would make the
industry, which depended on close „real-time‟ coordination across a whole range of
processes, unmanageable. Those within and close to the industry were therefore unconvinced
by the practicality of this option, so the debate continued without any detailed plan being
produced within government.
As the 1992 election approached, the Government nevertheless became committed to
privatising British Rail in the next Parliament and was forced to put its emerging
restructuring proposals into the election manifesto. Therefore a sketchy proposal was pulled
together at the last minute and put into the manifesto, proposing the creation of a railway
infrastructure company, and a franchising authority which would run a competitive process
to franchise train service operations to private sector operators, with subsidy provided if
necessary, as well as enabling „open access‟ operators to initiate their own independent
services in competition with others.
Implementing the restructuring and privatisation
When the Conservatives won the election in May 1992 the Government was then faced with
the task of implementing this plan. Given the sketchiness of the manifesto proposals, a period
of reflection and consultation might have been expected, in order to think through and
develop the details, to test them in debate and achieve consensus behind a robust plan.
Instead a White Paper was quickly published in July 1992 that added only a little detail to the
manifesto proposals. None of the operational details of the radical new structure had been
defined, shown to be workable, or tested.
The Government therefore decided to proceed on the basis of faith that the scheme could be
made to work and that the details could be developed during the implementation process. A
similar course had previously been taken in the complex restructuring and privatisation of the
electricity supply industry. A Railways Bill was introduced into Parliament early in 1993. It
contained no detail because none had been decided. It was drafted in the form of enabling
legislation to allow the Government freedom of action to develop the scheme, through
secondary legislation, licences and contracts, as required. The lack of detail in the bill meant
that Parliament never had the opportunity to debate or to test the privatisation plan and how it
was proposed to operate.
Even so, the bill was incomplete and needed copious amendment, not least because the
Government was pushing through its agenda against effective opposition (including from its
own backbenchers and the Lords) empowered by the wafer-thin majority resulting from the
1992 election. Accordingly, a series of concessions had to be made, so that the statute, which
came out, was hardly recognisable as the bill, which went in.
The troubled passage of the enabling legislation in the period to the end of 1993 meant that
the operational processes needed for restructuring and privatisation all then had to be defined,
tested, made workable, and implemented within a three-year window before the next
election, expected in 1997, with a view to making them practically irreversible. Time was
never given to testing the proposals through rigorous analysis and inclusive debate before
implementation began. It was clear that this was a very ambitious programme to attempt in
5
such a short time frame and one that was fraught with risks. The late decision to privatise
Railtrack added further strains to the integrity of the process and left too little time to prepare
the company properly for sale.
The implementation work was carried out by working groups of civil servants, advisers and
senior British Rail staff, each looking in parallel at specific aspects of the proposals, but
without questioning the practicality of the whole scheme. Accordingly, there was no overall
„master plan‟ giving guidance and explanation of the role of each component of the structure,
how they would each operate, and how the components of the scheme would inter-relate with
each other. The only plan was a programme management chart showing the time when tasks
had to be completed. It is therefore not surprising that there were many weaknesses in the
detail of the structure and mechanisms that were designed by this flawed, „bottom-up,‟
incremental process.
Nevertheless, the privatisations were successfully achieved within the timescales set by the
Government, and, very quickly, the industry succeeded in improving performance, passenger
and freight traffic grew by 35% and 50% respectively in the first few years after privatisation
and, freed from public sector expenditure constraints, investment grew substantially.
The lost vision of a commercially driven railway
Despite its hurried implementation, the structure for the privatised railway inherited by the
new Labour Government in 1997 had a basic logic to it and initially it performed relatively
well despite of the problems of a new and complex structure. The commercial arrangements
operated through a contractual matrix, mediated by charges and performance incentive
regimes between the key components of the industry. The 25 passenger train operating
companies (TOCs) received revenue from passengers and fixed annual subsidy payments
from OPRAF (later to become the SRA). The level of subsidy was determined through
competitive bids submitted to win franchises with significant protection from on-rail
competition that would meet at least minimum service levels set by OPRAF. The TOCs paid
pre-agreed lease and maintenance charges to the three rolling stock companies (ROSCOs) for
the rolling stock they used, and regulated access charges to Railtrack for the use of
infrastructure. In order to keep Railtrack „lean,‟ the maintenance and renewal of the
infrastructure were contracted out to 13 companies formed from British Rail and sold to the
private sector. There were also freight train operating companies that did not receive
operating subsidy.
The original vision of competitive open access, that had influenced the choice of industry
structure, did not survive the Treasury‟s other need to minimise subsidy by protecting
franchise train companies‟ effective monopolies during the franchising process. This
recognition was, however, fudged by agreeing a temporary regime of „moderation of
competition‟ authorised by the Regulator who, nevertheless, had inherited the promotion of
competition as a statutory objective from the 1993 Act. However, it was hoped that there was
sufficient competitive pressure in the train service franchising process and between the
ROSCOs and the infrastructure maintenance and renewal companies, reinforced by
contractual incentives, to make the industry efficient and responsive to market needs.
The initial results of these arrangements produced encouraging results. The outcome of the
franchising process offered the prospect of a rapid fall in the level of government subsidy
6
levels from £2.1 billion in 1996/97 to £0.9 billion in 2002/03. With rail traffic growing
rapidly and rising levels of investment, it was hoped that the Government would gradually be
able to disengage from most of the rail system and allow the private sector to take over the
risks and invest in a greater share of the future development of the industry. The need for
regional railways services to continue with public sector specification and support had,
however, remained an accepted inevitability throughout the pre and post-privatisation period.
There were, however, a number of flaws in the arrangements. Now, seven years after
privatisation, there remains an urgent need for these flaws to be corrected to enable the
contractual arrangements for the restructured and privatised railways to operate effectively,
within a coherent policy and strategic framework, on the basis of commercial incentives with
the minimum direct involvement of government. What is not needed, and would be most
damaging for the industry, is another round of major restructuring or, worse, any further
attempt to renationalise the industry. To date, the Government‟s initiatives and ad hoc
interventions have generally added further confusion to the contractual and incentive
framework for the industry, increased costs, and have moved the industry towards re-
nationalisation by shifting the risks in the industry away from industry operators and their
customers and back to the taxpayer.
Optimistic franchise bidding assumptions
One of the flaws in the original privatisation was that the TOCs‟ bidding assumptions for fare
box revenue and controllable costs proved to be unrealistic. The franchises were awarded on
the basis of the lowest subsidy bid and, as suspected by many commentators at the time, the
winning bids of some of the TOCs were based on optimistic assumptions. They were also
based on levels of infrastructure access charges, which were set by the Regulator, but which
were inadequate to fund Railtrack‟s requirements (see below). As a result, despite the
growth in traffic demand rail revenue hardly increased and many of the TOCs have
experienced financial difficulties and have required additional financial assistance from the
Strategic Rail Authority. Some of the franchises have been renegotiated with increased
subsidy, while others have been converted into short term management contracts, with
limited risk on the private operators, until they can be re-bid.
Escalating railway infrastructure costs
There have also been huge increases in infrastructure costs. When the Regulator set the level
of Railtrack‟s initial access charges in 1994, he was expecting the costs of maintenance and
renewal (M&R) of the infrastructure to be below £2 billion per year, as he applied substantial
reductions to the proposals put forward by the company (based on BR budgets and accounts).
In fact, Railtrack spent £2.5 billion per year on this task, funded by increased private sector
borrowings, and yet it was still heavily criticised by the Labour Government and the new
Regulator for under-spending. As a result, in the next regulatory settlement reached in 2000,
Railtrack was to be allowed over £3 billion a year to spend on M&R.
In the event, the Government put Railtrack into administration as a result of a cash crisis
resulting from the emergency expenditure after the accident at Hatfield in October 2000 and
the Government‟s withdrawal from its April 2001 refinancing agreement with Railtrack. This
was triggered by the company‟s proposal to have its price-cap regulation temporarily
suspended. Instead, the Government decided that Railtrack‟s role should be taken over by a
„not for dividend ‟ company funded from private sector loan finance, with supporting
guarantees by the SRA and comfort from the government. This has effectively returned the
7
risks of the infrastructure back to the taxpayer and, as observed by the Regulator, has created
difficulties in developing effective means of incentivising the management of Network Rail.
The accident at Hatfield precipitated a loss of control of railway infrastructure costs as
railway managers became risk averse (see below). Network Rail submitted a business plan to
the Regulator proposing a huge increase in expenditure to £6 billion per year for the next 5
years. And it has proceeded to spend at broadly this rate over the period 2001 to 2004 when
price controls were suspended, while the Regulator carried out a further review of
infrastructure expenditure requirements post-Hatfield. The Regulator concluded that Network
Rail should be allowed approximately £5 billion per year, compared to the £3 billion he had
earlier agreed for Railtrack. Therefore instead of subsidy steadily reducing while rail traffic
expands, as originally envisaged, the Government now faces the prospect of escalating levels
of subsidy, even beyond present levels, simply to maintain the current level of traffic on the
railway.
At the root of the difficulties experienced by Railtrack in managing its costs were fatal flaws
in the contractual arrangements that were made for the maintenance and renewal of
infrastructure during the privatisation process, and the subsequent failure to manage its
contractors effectively. The Government of the time laid down the policy framework
requiring outsourcing of M&R work and the detailed contractual arrangements were
designed by BR staff, most of who could be seen as having an interest in the outcome, since
they expected to be transferred to the new M&R companies. The maintenance contracts
created were fixed price and output-based, with performance regimes that had liabilities
which were capped in order to make the new companies more saleable. This effectively
meant that the maintenance companies themselves decided what work needed to be done and
made proposals on whether renewal rather than maintenance was required. Since renewals
were carried out by separate companies, the incentive was for the maintenance companies to
do as little as possible for their fixed price work and to seek to shift work onto the renewals
companies who were paid on the basis of work done.
Although it was paying for the maintenance and renewals concerned, and it had the legal
responsibility for the condition and safety of its infrastructure, Railtrack‟s day-to-day role in
M&R processes was limited. Since most of BR‟s engineering staff and much of the
information on the inventory and condition of its assets were handed over to the M&R
companies during restructuring, Railtrack did not inherit the means to manage its own assets
adequately. It did not have the tools to either monitor the work that had been done, or to
decide what was needed or to determine the costs of maintaining and renewing its own
infrastructure. It relied on its limited inspection resources and indirect „output‟ indicators,
such as the incidence of broken rails, to assess trends in the condition of its assets. Railtrack
could not therefore be sure it was receiving value from the increasing amount of money
being spent on M&R, or whether some of it was being lost to inefficiency or excess profits
for the M&R companies.
Though they were largely not of its initial making, Railtrack was slow to address these
inherited issues. It did not manage its contractors effectively and it allowed them to withhold
information and neglect to keep the asset data bases up to date which were in their charge.
Although it made some changes in its organisation and supply contracts, it did not begin to
develop its own asset register until the year or so before it was put into administration and,
because of agreements made at the time of the sale of the M&R companies, it was not able to
8
restructure its maintenance contracts until 1998 and it did not build its engineering expertise
sufficiently.
One consequence of the continuing lack of access to essential information on infrastructure
assets was that neither Railtrack, the Regulator, the SRA, nor any other interested parties had
the means to understand the efficient costs required for maintenance, renewal or investment
in the railway infrastructure and therefore how to manage these costs, or to forecast them
reliably. This made the application of conventional price regulation, through RPI-X controls,
problematic since such a framework depends on reliable forecasts of efficient expenditure
requirements. Thus, although there was a long drawn out regulatory review process carried
out to reach a determination on the costs required, both Railtrack and the Regulator (and
subsequently Network Rail) have been working in the dark throughout this process with
inadequate information. There therefore remains a high level of uncertainty surrounding the
cost requirements of the railway, while the incentives to manage these costs have been
considerably weakened by the creation of Network Rail, with many more of the associated
risks now lying with the Government.
The decision of Network Rail to bring some infrastructure maintenance under its direct
control may help it understand maintenance requirements and its costs better in the short
term, but it should not be the long-term solution. There is no reason why railway
infrastructure maintenance and renewal cannot be successfully outsourced from the asset
owners. Other safety critical industries, such as the airlines, have done this efficiently, and
indeed railways throughout Europe (including France) have been moving in this direction. It
has been shown to be practicable for not-for-profit asset owners to outsource in this way, as
evidenced in the water industry by Glas, one of the models for Network Rail. What is needed
is for it to done properly in the railways. As soon as Network Rail develops its asset register
and populates it with data, it will be better able to manage its assets and outsource
maintenance and renewal efficiently without losing control.
Other contractual flaws
Although they worked better, there were some flaws in other contracts designed during the
implementation process for privatisation. The structural arrangements and divided
responsibilities in the industry have exacerbated these weaknesses and contributed to the
difficulties in addressing them. Each set of contracts - the franchising agreements between
the TOCs and the SRA, the access agreements with Railtrack, the leasing agreements with
the ROSCOs and the infrastructure M&R contracts - were designed separately by different
bodies, they run for different time periods and their performance regimes were not consistent
with one another. Consequently, despite efforts by the Regulator to improve the contractual
arrangements, there is still no systematic consistency between the incentives operating on the
interdependent operating organisations in the industry, so that they frequently work against
each other. What is urgently required is to build on and implement the work already carried
out by the Regulator to develop a revised set of contracts between the TOCs, ROSCOs,
Network Rail and the infrastructure M&R organisations, which are „back-to-back‟ so that
each has consistent and self-reinforcing incentives in their performance regimes.
Thus, as an example of the problem, the Railtrack performance regime gave it a strong
financial incentive to improve train delay performance and hence it delivered a 40%
improvement in train delays attributable to it in the first two years. But Railtrack was given
little incentive to accommodate growth in demand on the network since most of its access
charge revenues were fixed. The TOCs, on the other hand, had a strong incentive to grow
9
demand and revenue, but relatively weak train performance incentives. Hence passenger
demand and the number of train services grew rapidly but the TOC‟s performance in
minimising train delays was disappointing. This growth, however, put further strains on
Railtrack‟s under-funded infrastructure and neither the Government nor the SRA has
developed policy leadership to address the consequential mismatch between supply and
demand. Although attempts were made by Railtrack and some TOCs to negotiate increases in
access charges to pay for additional capacity, this ad hoc approach failed at the network level
because the SRA was unwilling to become involved in agreements which would inevitably
result in an increase in subsidy requirements
This example, of course, also illustrates the power of financial incentives to drive
performance and the positive results that can be achieved (at least when such incentives are
operating on private sector companies). This should have pointed the way to improving the
incentive framework for the industry to address the weaknesses inherited from the
privatisation process. Unfortunately, the present Government has pointed the industry in the
opposite direction and has failed to address the weaknesses in the incentive regime, while
undermining the basis on which the existing incentive structure operates.
Labour’s response
Labour in Opposition was consistently hostile to the principle of privatising of British
Railways. Various representatives of the Opposition made threats of re-nationalisation during
the process of privatisation that blighted its progress and resulted in poor value for money
from some of the sales. The rolling stock companies, in particular, were sold off too cheaply.
Despite the efforts of the sales teams, the value of the guaranteed cash flow from leases for
the rolling stock transferred to the private sector and the potential value of new leases did not
convince the markets sufficiently to offset the perceived political risks, with the result that
the companies were subsequently re-sold at a great profit to the original private sector buyers
when these perceived risks dissipated.
The Labour Party‟s disengagement from the details of the privatisation process meant that it
had not thought through its policy to put right what it eventually inherited. It was soon clear
that its threats of outright re-nationalisation would be hollow because of the costs involved.
Instead its policy on taking office amounted to little more than a slogan „to restore a publicly
accountable railway by tough regulation.‟ It lacked any thought-through detail of the
consequences of this policy for incentives and performance in the industry. Given the
weaknesses highlighted above, as well as the successes that had been achieved by private
sector incentives, it would have been sensible for the new Government to have carried out an
early careful review of the structure and operation of the contracts and mechanisms in the
industry, drawing on the expertise and detailed knowledge which had been accumulated
during implementation. This would have enabled the Government to address the problems it
had inherited through selective but consistent changes. Instead it decided that what was
needed was „tough‟ regulation and. „strong‟ leadership to force the industry to improve.
No strategic leadership and confused regulatory arrangements
In fact, the Labour Government inherited a rather confused set of regulatory arrangements,
which it proceeded to confuse further. One of the major mistakes of the Conservative
Government was not to clarify the role and specific responsibilities of the two regulators, a
10
failure that led to endless friction. This was not helped by the Labour Government appointing
two unusually adversarial people to the relevant posts.
The role of the franchising authority, OPRAF, was to let the train operating franchises
against minimum service specifications so as to achieve value for money from the subsidy
provided. The original idea had been that the first franchises would run for a limited period
of seven years to allow time for the train operating companies, Railtrack and OPRAF to
determine the investment needed for future development. This investment would then be
funded during the next round of franchises. Railtrack was funded to carry out its M&R work
but not to expand capacity substantially since long run trends in rail usage had indicated that
there would be little demand growth. Any additional capacity that might be needed before the
next re-franchising round would be agreed by ad hoc negotiations between Railtrack and the
TOC concerned, with the support from OPRAF, if and when justified economically
In fact, the unexpected growth in demand did put strains on infrastructure capacity, but, as
noted above, the process of negotiating capacity enhancement bilaterally between Railtrack
and the TOCs - from which, in the event, OPRAF/SRA largely stood back - did not prove
workable, partly because any enhancement investment in a loss-making industry was
unlikely to be financially viable and would require substantial support from government. It
seemed that OPRAF/SRA, as the arbiter of economic value and the primary funder of
investment, would need to take a more central role if robust capacity enhancements were to
be agreed.
The significant exceptions to this pattern were the West Coast Mainline (WCML) renewal
and upgrade project and the Thameslink 2000 project, both of which had been well advanced
in the planning stages within BR.
In the case of the WCML project, Railtrack first entered into a fixed price, output-based,
access contract with Virgin that had been negotiated between OPRAF and Railtrack before
the franchise was awarded. Following bilateral commercial negotiations, this was
subsequently amended to include further upgrades specified by Virgin, which itself was then
further altered to include (unfunded) public sector requirements specified during the
regulatory approvals process. In addition to the extremely onerous commercial terms
involved in the resulting deal, Railtrack‟s experience with attempting to deliver its
commitments for this project exposed its lack of understanding of its existing assets, the
proposed new (and untested) signalling technologies, its inability to plan robust timetables
for the new service commitments, and its inability to estimate or manage its costs. The
estimates for completion escalated from £2.1 billion to £13 billion at one stage, before the
scope of the project was subsequently cut back by the SRA, to enable the current estimates of
about £10 billion to be derived. This loss of control of its flagship project was a major cause
of the subsequent failure of the company.
The Strategic Rail Authority (SRA) was created to replace OPRAF so as to provide strong
strategic leadership in the industry. It was a recognition of the difficulty being experienced in
dealing with the strategic and policy issues in the industry, such as long term investment, the
balance between subsidy and fare box revenue, efficient targeting of subsidy and the
management of the supply/demand balance for rail capacity.
However, once again the revised role, responsibilities and mode of operation of the SRA did
not appear to have been properly thought through or communicated. Instead of developing a
11
strategic plan to address the policy issues and choices for government, and preparing the
investment plan for the industry, the new Chairman decided that the process of developing
the railway should be driven through the train operators, who would submit their own plans
in competitive bids for long term franchises, with investment commitments, but without the
aid of guidance, or useful criteria, from the SRA on what it would wish to buy as the primary
funder for most of the investment.
This created considerable difficulties for the bidders in trying to anticipate what scale and
composition of investment would be attractive and socially worthwhile, given that most
investment could not be profitable and would only be justified on economic/social criteria. It
also created difficulties for the SRA in trying to compare the relative merits of very different
alternative plans and bids through a competitive procurement process under public sector
rules. The problems were further compounded by the slow development of revised
economic/social appraisal criteria within the SRA, which were founded on the New
Approach to Transport Appraisal (NATA) that relied on multi-criteria evaluation. The SRA
appears to have made little progress in developing a strategy for the development of the
railway and only recently have appraisal criteria for evaluating the economic viability of rail
services and investment been finalised.
These uncertainties, together with the inherent difficulties in making reliable estimates of the
costs of major investment in the railways, made the whole process extremely risky for
bidders. Railtrack, the owner of the infrastructure, was marginalized in this specification
process. In consequence, because critical projects such as the connection of the high-speed
Channel Tunnel Rail Link to the conventional network, and the East Coast Upgrade, had to
proceed with planning and initial works, Railtrack effectively planned its enhancements in
parallel. To meet licence requirements of the Regulator, it published a series of Network
Management Statements with increasing lists of (unfunded) investment proposals which
SRA failed to prioritise.
The result of these parallel processes was an inability of Railtrack and the SRA to agree clear
project specifications and firm costings for them. Railtrack cost estimates suffered from a
systematic tendency to creep up, as had British Rail‟s. Coupled with its West Coast Main
Line problems, this experience convinced the SRA that Railtrack did not have the
management capacity to implement major new investment schemes. Therefore the SRA
proposed that „special purpose‟ vehicles be used to raise the finance for the schemes and that
private sector consortia with the relevant skills would manage them. It was unclear how the
complex interfaces and interdependencies within the network would be managed by this
means. In the event, no such schemes were developed to a successful conclusion before
Railtrack was put into administration, and two years and a lot of resources were wasted.
Little work has been done to evaluate whether the present levels of infrastructure
expenditure, recently sanctioned by the Regulator in his Interim Review of Network Rail‟s
expenditure plans, are economically justified. The SRA‟s Network Output Statement appears
to imply that it views different outputs to those proposed by the Regulator as being more
economic, but as the Regulator did not base his conclusions on this Statement, the true
position is not clear. Without these essential elements, it is hard to see how the SRA can
perform its role of providing strategic leadership. The lack of a rigorous public sector process
for planning, evaluating and prioritising rail investment using economic appraisal criteria has
left the industry in some confusion and unclear over the long term direction and future
12
development. In the meantime, the pragmatic response has been to try to maintain services
within the financial resources currently available.
The confusion has been compounded by the role of the Regulator. The independent Rail
Regulator was created to provide comfort to private investors in TOCs that the government,
through OPRAF, would not set both the quantity of outputs required from TOCs in
franchises and also the price of their inputs in monopoly supply (in access agreements). It
also had a role in consumer protection and carried out tasks related to this objective with the
support of passengers‟ committees.
The original role envisaged for the Regulator was to establish the level of Railtrack‟s access
charges, to review and approve the access agreements between Railtrack and each of the
TOCs and to oversee the operation of the contractual arrangements in the industry and
resolve any disputes arising. However, the Labour Government‟s commitment to „tough‟
regulation soon led to the introduction by the Regulator of additional, and overlapping, direct
licence based controls on Railtrack, in addition to its commercial obligations through its
TOC contracts. These licence obligations enabled the Regulator to require Railtrack to invest
and to meet output targets decided by the Regulator, not the SRA, although the SRA would
effectively be required to pay for these investments through government subsidy. These
licence based obligations resulted in responsibility for determining Railtrack‟s investment
obligations being split between the Regulator and the SRA with no clear demarcation of
responsibilities.
The disastrous failure of the SRA to develop a strategy and an investment plan for the
industry left a void which the Regulator sought partially to fill, but without having the
essential information, tools and expertise. He relied heavily on consultants to carry out the
technical work for his reviews, although these advisers often had insufficient reliable
information to work from and they were unable to make the policy judgements implicitly
required for their work. It should have been obvious that the idea of one body, the SRA,
controlling the specification of train services and funding the costs involved, while another,
the Office of the Rail Regulator (ORR), separately determined the outputs, investment
obligations and funding requirements of the infrastructure paid for by the train operators, was
a nonsense.
Since the government, and not the rail user, is the marginal funder of the rail infrastructure,
the effect has been that the Regulator‟s decisions on the volume, quality and price of
infrastructure outputs effectively commits the government to funding the consequences of
those decisions, although it is the SRA, as the agent of government, which has the budgetary
responsibility for the railway. It is difficult to see what the Regulator „having regard to the
SRA‟s budget,‟ as he is required to do, can mean in practice if the SRA has not developed a
properly evaluated and prioritised plan for the development of the railway and the Regulator
seeks to impose his own output targets on the infrastructure owner (essentially based on
historic outcomes) without the necessary analysis for their justification.
Given the volatility demonstrated in the cost estimates prepared by the infrastructure owner
and accepted by the Regulator in recent years, and the inherent uncertainty in any estimates
of these costs under current circumstances, it is unlikely that the Government would want to
accept the levels of infrastructure expenditure currently proposed by the Regulator until
further work has been done to understand the reasons for the massive cost escalation in the
13
management of the rail infrastructure and whether the latest levels of funding can be justified
economically.
Responding to risk
There is little doubt that one of the major causes of infrastructure cost escalation since 2000
has been the direct and indirect consequences of the heightened level of risk aversion that
now permeates the behaviour of managers, staff and regulators operating in the industry. The
reason for this can be traced to the reaction of the Government and the media to major
accidents on the railways since privatisation, but it was also sensitised by the hostility
exhibited by the Government (and its appointees) to the privatised industry from the outset.
This created a climate of blame and recrimination and a corresponding defensiveness in the
industry that was not conducive to efficient operations.
Early on John Prescott mounted a campaign of public criticism of the privatised railway for
its performance record. This was later reinforced by regular public criticism from the new
Regulator, focussed largely on Railtrack. When the Ladbroke Grove accident occurred in
1999 the reaction of the Government was to „spin‟ the blame onto Railtrack, notwithstanding
the fact that the accident was caused by a driver passing a red light at danger. In fact, the
safety record of the privatised railway had been better than under BR and yet politicians and
the media created a crisis atmosphere following Ladbroke Grove. The police turned the site
of the accident into a crime scene, John Prescott branded the industry a national disgrace and
Railtrack was to be „stripped‟ of its safety responsibilities. Safety would be improved
„whatever the costs‟- a statement that was a considerable hostage to fortune. Railtrack was
vilified in the media, the home addresses of Board members were displayed in the press and
journalists harassed their families, as well as those of drivers and signalmen.
The safety regulators and the Cullen inquiry that followed drew the conclusion that there was
a very low public tolerance of rail accidents. Consequently, it was judged that expenditure to
improve safety was justified, even though the money would have a much lower impact on
safety than if spent elsewhere. According to Professor Andrew Evans, the average cost of a
life saved by the installation of TPWS is £11 million. Safety constraints became a major
impediment to progressing investment in the railways. The safety regulator operates
independently from the other two regulators and the impact of safety expenditure is not
subject to economic evaluation to enable policymakers to make the relevant trade-offs.
Furthermore, the behaviour of those working in the industry became excessively cautious
under the threat of jail sentences for railway staff and managers.
Consequently, when the accident at Hatfield occurred in October 2000, and it was quickly
apparent that the cause was the breakage of a badly deteriorated rail, Railtrack management
panicked and imposed hundreds of speed restrictions throughout the network wherever there
was any suspicion of the condition of the rail. The damage to rail services was immense and
the impacts on costs within the industry are still being felt. Without public support, rail
managers were no longer willing to make the sensible judgements about levels of risk that
are essential to efficient railway operations.
14
The accident at Hatfield and the response to it, appears to have precipitated a total loss of
control of costs and train performance in the industry. Before Hatfield, Pollitt and Smith1
have demonstrated that operating costs in the railways were being managed reasonably
tightly, although cost estimates of major projects were escalating and there were doubts over
whether maintenance activities were being managed effectively. The enquiry into the rail
breakage at Hatfield confirmed those doubts. Since Hatfield, engineers and safety regulators
have sanctioned huge expenditure increases, and are proposing much more, simply to restore
the levels of performance prior to the accident by the end of the decade. Alistair Morton‟s
characterisation of the industry‟s reaction to Hatfield as a „nervous breakdown‟ was accurate;
and it still has not recovered. It will be difficult for those in the industry, including safety
regulators, to return to rational behaviour unless they are given greater support from
politicians and the media. In the meantime, the industry cannot deliver expected performance
at a sensible level of costs until confidence is restored to those working in the industry.
What is to be done?
In our view, the Government should avoid the temptation to impose yet more radical change
on the industry that would result in further destabilisation of those engaged in providing train
services to passengers. However, some changes are required in the way policy is determined
and is translated through the regulatory structure of the industry and the way that the
incentive arrangements operate. There are three priorities for change:
The institutional arrangements for regulation, policymaking and planning in the
industry must be simplified and streamlined so that decisions can be taken quickly
and consistently on the basis of sound analysis. The conflicts and duplication of effort
inherent in the current arrangements must be addressed.
The industry must operate under a system of contracts based on a consistent structure
of incentives which can be allowed to operate without ad hoc interventions that
undermine confidence and increase risks and costs in the industry
The industry must recover its confidence and develop a systematic approach to
prioritising and determining the appropriate level of safety expenditure to maintain
the good safety record of the railways. And the climate of risk aversion must be
changed. To achieve this, politicians and the media should change the tone of their
response to the railways and promote rational public debate about the industry.
A new role for the SRA
The Strategic Rail Authority should take the central leadership role in the industry. The SRA
should carry out its strategic planning role much more effectively in the future and develop,
with government, the investment plan for the industry on the basis of a coherent strategy and
consistent policies. It should cease involving itself in the direct management of the industry.
The main change in SRA‟s role would be to give it a new direct commercial relationship
with Network Rail. In addition to negotiating the train operating franchises, the SRA would
contract the complementary infrastructure capacity supply agreements directly with Network
1
The Restructuring and Privatisation of British Rail: Was it really that bad? M G Pollitt and A S Smith
Institute for Fiscal Studies 2002
15
Rail, including performance incentives, and it would pay charges for outputs delivered by
Network Rail.
The Regulator should no longer have the role of setting Network Rail‟s expenditure plan or
deciding infrastructure outputs. The direct role of the Regulator in decision-making in the
industry should be reduced in consequence, but it would have an important role as a
mediating and appeals body to provide the necessary protection to private sector investors.
Its expertise and knowledge of the railways could also be utilised in carrying out periodic
reviews of the SRA‟s performance. Its role as an appeals body and in reviewing the
performance and effectiveness of the SRA would provide a powerful check on the SRA and
essential accountability for how it carries out its vital functions.
This new arrangement would remove the confusion, conflict and duplication of functions that
have blighted the industry for the past 7 years. Establishing a direct commercial relationship
between the SRA and Network Rail would provide the clarity which is lacking in the present
arrangements of overlapping contractual incentives, licence obligations and indirect policy
guidance governing Network Rail‟s activities. In time, it should also enable a more informed
discussion over the level of infrastructure costs and the factors affecting efficiency than is
possible through a regulatory review process.
The current Regulator has argued that independent regulation of Network Rail is needed to
protect private investors. However, Network Rail‟s debt funding is already effectively
guaranteed by the SRA (and behind it, the government), so that investors are protected. The
SRA therefore has no interest in driving an unreasonable bargain that threatens Network
Rail‟s ability to service the debt, which the SRA is itself supporting. Furthermore, the
application of conventional RPI-X price controls on rail infrastructure supply has been
shown to be ineffective, since it is not yet possible to forecast required expenditure to deliver
defined infrastructure outputs with any reliability. The Interim Review made further
concessions to recognise this reality, with much of the next five year‟s expenditure/output
mix now being subject to further review. To secure funding confidence, some “emerging
costs” have been replaced by “emerging outputs” in the regulatory contract. This lack of
transparency and certainty over the present and future efficient costs of supplying rail
infrastructure places a major question over the effectiveness of the management incentive
framework that has been put in place for Network Rail and over the Government‟s decision
to create this quasi-nationalised organisation.
The original concept that the rail infrastructure provider would be driven by commercial
contracts with its users, the TOCs, with the Regulator setting price caps every five years, has
therefore not been effective. The TOCs could not commit funding for infrastructure
development without the involvement of the SRA and they could not easily take account of
the investment needs of other TOCs that may be affected by the schemes they were
promoting.
Since the SRA authorises the franchises to provide train services and funds the non-
commercial elements of these services, it should also underwrite the purchase of the essential
infrastructure to enable these services. In the process of doing this, it would need to plan the
infrastructure investment required, working with Network Rail and the relevant TOCs, to
agree the level of infrastructure outputs and the associated performance measures on which
Network Rail would be rewarded. These incentives would be tailored to take account of the
level of uncertainty over forecast expenditure and outputs, and would give greater focus to
16
efficient delivery than is possible under a regulatory review process where the Regulator is
obliged, under statute, to balance a whole range of public and private sector interests and
objectives. Seeking such a balance in a publicly funded, privatised industry will always risk
replacing the highest common factor by the lowest common denominator.
The implication of this arrangement is that Network Rail would earn most of its revenue
through charges paid directly by the SRA rather than through the access charges paid by the
TOCs. To an extent the switch to direct payment by the SRA for infrastructure capacity
reflects what is already happening now, since the Regulator acceded in 2000 to a request that
40% of Network Rail‟s income should be paid by grants from the SRA, rather than through
access charges and the SRA has requested that the proportion increases following the Interim
Review. However, the lump sum grant currently envisaged would not be structured to
provide any incentives on Network Rail to deliver what the SRA is paying for – indeed the
Regulator has made it clear that such a grant must be indistinguishable from access charge
revenue for accounting purposes.
Rather than paying Network Rail through a lump sum grant that the Regulator resets
periodically as an outcome of access charge reviews, we would propose that the SRA would
pay disaggregated fees for individual lines which would be structured to include incentives
on Network Rail to deliver specific outputs for those lines, as the access charge and licence
regime does now. These fees would pay for the longer-term stewardship work and outputs
that the SRA requires of Network Rail, regardless of the changing allocations of access rights
under successive refranchising initiatives. There would be a performance regime in the
contract between the SRA and Network Rail with penalties and rewards for under and over-
performance based on performance measures relevant to the delivery of contracted capacity.
We would envisage that this performance regime would be developed in consultation with
the TOCs so that the performance measures used are relevant to the reliable long term
delivery of infrastructure capacity.
The TOCs would continue to have their contractual relationship with Network Rail through
access agreements as now. However, the TOCs would only pay the variable element of
access charges designed to cover Network Rail‟s variable costs, as freight operators already
do and in line with emerging european practice. The contract between the TOCs and
Network Rail would be concerned with day-to-day operational performance, including the
availability of infrastructure, train delays and customer service requirements. Performance
regimes between the TOCs and Network Rail would therefore continue to incentivise day-to-
day operational performance.
Clearly the two performance regimes between Network Rail and the SRA and the TOCs
must be designed to ensure that Network Rail faces a consistent set of incentives from its two
customer bases. This need for complementary incentives with the public and private sectors
is already embedded in the licence and access contract currently in place with Network Rail:
the essence of our proposal is that, in each case, it would now be the respective payers of the
„piper‟ who would call the tunes concerned.
There may nevertheless be a continuing risk that confusion and tensions could arise in this
dual contracting arrangement, which splits Network Rail‟s responsibilities between two
customer types, one buying longer term network capacity and capability and the other day-
to-day performance. For instance, the TOCs may seek to blame their non-performance on
deficiencies in the infrastructure outputs agreed or supplied between Network Rail and SRA.
17
Such tensions are (and always have been) at the heart of a railway on which commercial and
public services run alongside each other.
The contractual structure and performance measures used must therefore be designed with
great care to mitigate this risk and to ensure that performance incentives are consistent and
understood and agreed by all parties from the outset. In practice, it will take time to align all
the incentives in the contracts since this task will have to be carried out on a rolling basis as
the franchises come up for re-bidding or renegotiation.
Clearly a single direct contract between each TOC and Network Rail would be the ideal
option if the TOCs were not dependent on government subsidy. But the direct involvement of
the SRA is inevitable while the railway is so heavily dependant on government subsidy and
while there is such a high level of uncertainty over longer term infrastructure costs and
outputs. In our view, the arrangements proposed here are far preferable to the current
arrangement where the Regulator makes - and remakes - the decisions on Network Rail‟s
outputs and expenditure without the essential knowledge of industry requirements or the
public expenditure trade-offs required to ensure the most effective and affordable solution.
Development of a strategy for the railway
It is the role of government to make the major policy decisions that will provide the
framework for a strategy for the railways. These key policy decisions include the balance of
resources devoted to road and rail transport, the financial limits on rail subsidy, the balance
of rail funding between subsidy and farebox revenue, setting criteria for the use rail subsidy
and endorsing the consequential decisions on the regulation of fares, the size of the rail
network and the scope of rail services supported by subsidy.
It is the role of the SRA to provide government with relevant analysis and advice on options
to enable it to make these key policy decisions. The SRA has available the essential
information on rail markets and costs to carry out this kind of policy analysis. It should also
use this information to develop a strategy and investment plan for the railway based on the
framework of policy decisions provided by government outlined above. The strategy and
investment plan will enable the SRA to become an informed and rational purchaser of
coordinated infrastructure and franchise outputs, something that the Regulator is not in a
position to achieve. This strategy needs to be more than simply a shopping list of projects. To
summarise, the strategy will need three key elements:
The Government will work with the SRA to develop policy guidance on the balance
of funding between road and rail and the consequential financial limits on rail
subsidy. Criteria for determining the allocation of rail subsidy should be developed
together with a consistent policy on the regulation of rail fares to ensure that the
appropriate balance of funding from rail users and taxpayers is achieved.
Based on this policy guidance from Government, the SRA should review the
allocation of subsidy between services across the network, and the scope and scale of
services supported by subsidy, to ensure that limited taxpayers‟ funds are being used
in the most effective way possible to achieve the stated goals of subsidy. It should
publish the investment criteria it uses and the calculations that show its use of them.
Once a clear policy on pricing, subsidy and the scale and scope of services required is
developed, the SRA should then develop a coherent plan for the provision and use of
rail capacity, the investment required and the level of funding.
18
Neither British Rail nor Railtrack were ever given this sort of rationale policy framework on
which to plan rail services and investment. If the railway is going to move towards delivering
the Government‟s policy of long term growth in rail transport usage at affordable costs, the
Government must give it a coherent and forward looking policy framework which should be
published through the Directions and Guidance given to the SRA. It is the Government
which must make the key strategic policy choices and therefore the Government must set out
the policy agenda for the SRA and give it clear guidance on the issues it wants addressing. It
is the SRA‟s job to address this policy agenda, to carry out the analysis and present the
policy choices and their consequences to Government. In practice, there is a spectrum of
policy choices that must be made, from high-level issues to detailed judgements on
operational matters. It is important that the Government restricts its guidance to providing the
high-level policy framework and delegates decisions on operational matters with policy
implications to the SRA without intervening. Otherwise it will undermine the SRA‟s
authority and effectiveness.
One of the most important policy issues is to attain the appropriate balance between revenue
from customers and subsidy from taxpayers for the funding of the railway. This requires an
evaluation of the wider economic and social value of subsidy on individual lines- principally
any net environmental benefits of reduced road usage and the benefits of reduced road
congestion. It also requires an analysis of the level of passenger fares that would achieve a
balance between supply and demand for available capacity.
The growth in rail traffic which has been experienced in recent years has been due, in part, to
a fall in the real cost of rail travel relative to road, as roads have become more congested.
Marginal traffic growth is desirable if it fills under-utilised capacity. But at the point when
major investment is needed to cater for further increases in demand, the most economically
viable policy is usually to delay the need for investment by increasing charges to restrain
demand, until such time as the level of potential revenue and external benefits meets the
relevant appraisal criteria.
There is no case for encouraging additional demand on severely over-crowded services by
subsidising fares. If Network Rail‟s underlying efficient costs really are at, or indeed much
more than half their current level, it is unlikely that much, if any, expansion in capacity is
either affordable, or economically justified. Indeed, at the current level of costs of the rail
infrastructure, the scale of the existing network will need to be reviewed and cut back to
make it affordable. However, to date, the Government has not given the SRA any policy
guidance to ask it to carry out the required analysis to make rational decisions on either
investment or pricing/subsidy policies. Without a coherent policy framework within which to
work the SRA has concentrated on tactical measures to achieve a better balance of supply
and demand within its financial budget through timetable changes.
Thus the second, and related, policy issue that should be addressed is the size of the network
and the extent of rail services. The distribution of subsidy between the various services
across the railway network has never been rational. This is largely because the issue of
closure or withdrawal of rail services has been deemed to be highly sensitive by politicians
and the media, and hence has generally been ducked by those responsible for advising
governments on transport policy. Yet this failure has resulted in resources being dispersed
ineffectually around the network and has prevented the passenger railway from concentrating
on the markets where it can win against road transport (i.e. mainly intercity and commuter
19
services in London and other conurbations), and hence make the greatest environmental and
economic contribution. It has thereby weakened the railways‟ long-term position and
viability.
It is unclear whether the supposed sensitivity of this issue reflects the realities of public
opinion, if faced with the real costs and the choices involved. Any discussion of the issue of
rail closures quickly becomes highjacked by user groups (who very often represent potential
rather than actual users), and the media coverage has tended to reflect this audience rather
than promote rational debate. Politicians often under-estimate the public‟s ability to accept
rational policies, if properly presented, as has been demonstrated by the acceptance of the
London congestion charge against dire warnings. The Government and the SRA should not
duck this debate. The escalating costs of the railway only emphasise the need to analyse
carefully the relative merits of the rail services across the network so that available resources
can be allocated efficiently and not simply to maintain the status quo.
Attaining clarity over these critical policy issues - the amount and distribution of subsidy, the
size of the network, the extent of rail services and the means to match supply and demand
using most effective price/subsidy policy - is essential for the SRA to develop a rational
investment strategy for the railway, to specify the train operating franchise requirements it is
prepared to pay for, and to specify the outputs that it requires from Network Rail. Only then
will it be able to evaluate alternative projects and prioritise investment properly. The SRA
will need to work closely with Network Rail which, as infrastructure owner, should have the
necessary detailed knowledge of its infrastructure condition and requirements to propose
infrastructure investment options, which the SRA will need to prioritise and decide. By this
means the SRA can present a credible overall investment plan to government setting out the
implications of alternative levels of subsidy support and pricing and investment policies.
Regional devolution
The Government indicated in its announcement an inclination to consider devolution of
decisions on support and the operation of train services. This is consistent with other policies
initiatives of the Government and has much to recommend it. Already responsibility has been
devolved for train services in Merseyside and the SRA recently announced that it would be
devolving responsibility for the planning, funding and operation of a number of rural
services, on the basis of vertical integration of both train operations and infrastructure. This
policy would enable the SRA to carry out its strategic role in planning the overall provision
of services and providing advice and expertise in franchising, while leaving the detailed
specification and operational management to local authorities. This should enable services to
be planned to meet local requirements and for engineering standards to be adjusted to match
local needs more efficiently. International experience of small-scale operation of „short line‟
services in the USA and local franchise operations in France and Switzerland and elsewhere
indicate the advantages of this policy.
We therefore generally endorse this initiative. However, there will be severe practical limits
to the extension of this policy. Where train operations are self-contained with little
interaction with other services, devolution can proceed without the need to negotiate affects
on other services. However, where different services share the use of the same infrastructure
there is considerable potential for conflicts over the allocation of costs, appropriate standards,
prioritisation in use of capacity and the shared funding of improvements. The addition of
another dimension to an already complex matrix structure of responsibilities would not be
appropriate. Therefore the case for devolution needs to be judged carefully on the merits of
20
the particular situation and must take account of the practical implications of the
interdependences and interfaces that exist within the railways in Britain. It is not a panacea
for overcoming the complexities of the current structure of the industry, but it can reduce the
scope of the problem in particular geographical areas.
Changing roles and relationships
The privatised railway has not had a good record for developing constructive relationships
between the various bodies concerned. This was partly because the transition from the
monolithic culture of BR to commercial relationships between a number of different
organisations, with overlapping interests, mediated by an unsatisfactory contractual matrix,
was rushed and proved difficult. This was not helped by the adversarial approach of some of
the key personalities involved. These mistakes must not be repeated.
Careful thought will be needed about the respective roles of the SRA, Network Rail, the
TOCs, the Government and the ORR and also about the way the process for planning and
delivering infrastructure expenditure and outputs should operate. These should then be
documented and agreed. In the past, relationships in the industry have been developed on an
ad hoc basis, often resulting in misunderstanding and conflict.
The success of these proposed new arrangements for running our railways will depend
heavily on the effectiveness of the SRA organisation. The SRA will need to develop a new
relationship with Network Rail as its customer. The working relationship between the SRA
and government will also need to be developed so that the Government sets the policy
agenda and makes the key policy decisions (expressed in its Directions and Guidance to the
SRA) and also provides political support to the industry, while the SRA carries out the policy
analysis and provides advice on policy options.
The SRA will need to „raise its game‟ and step back form attempting to micro-manage the
industry. One of the reasons that it has been drawn into detailed decision- making is that the
institutional structure of responsibilities and the framework of incentives which govern
behaviour in the industry was poorly designed. This has led some to call for a return to a
„command and control‟ approach to management that was said to characterise the old British
Rail. In fact, BR worked hard during the 1980‟s to change this culture by introducing sector
management and internal management incentives based on devolved responsibilities for costs
and revenues. An industry as complex as the railways can only be managed effectively
through devolved responsibility within a framework of consistent, self-reinforcing
incentives; and it is this that needs to be restored.
The tasks proposed for the SRA are not easy. They require clarity of vision, strong technical
and commercial skills and effective political and communication skills. The SRA will need
to win the support of the industry and the public to do this. And the Government must be
clear and consistent by setting the policy framework within which the SRA must operate and
refrain from other intervention in the industry.
New role for the Office of Rail Regulation (ORR)
The role of the ORR under these new arrangements for the industry would be to help make
the SRA accountable to consumers, the Government and Parliament, to provide protection to
private investors against unreasonable behaviour on the part of the SRA or monopoly abuse
by Network Rail and to protect the final consumer. Given the powerful role which the SRA
would have under these proposals, it is essential that it be held properly accountable for both
21
its decisions and its efficiency. As an agent of government the SRA should act in the public
interest. But there may be occasions when the inevitable conflicts and choices it must make
could result in abuse or grievances by the parties affected.
The ORR would stand outside these relationships and would no longer play a direct role in
decision taking in the industry. This would enable it to take an independent view of any
conflicts arising and to act as a mediating influence through the industry processes. It should
also act on behalf of the consumer in addressing any issues arising, either in response to
complaints or on the initiative of the ORR. The ORR should be involved during the process
of developing all the key contractual agreements in the industry. The Regulator
should review and, if necessary, comment on these agreements during their development. He
should stand ready to mediate if there are disputes arising between industry parties and, in
defined circumstances, act as an arbiter for disputes brought by industry parties. However,
we believe that the current requirement in legislation that the Regulator approves all access
agreements should be removed from the Railways Act, since it could be used by a regulator
to intervene directly in the SRA‟s negotiations with industry operators. The ORR‟s role
should be to facilitate agreement by mediation and, in the case of disputes arising, to arbitrate
between the parties. If there are serious disputes arising between parties that the ORR is not
able to resolve the parties can appeal to the Competition Commission which can decide
whether to hear the appeal if it is of sufficient importance to warrant it.
However, the ORR should be bound to act within the constraints of the finances available to
the industry and should no longer be empowered to make decisions requiring an increase in
government support. In addition, the ORR could provide an additional check on the
accountability of the SRA by carrying out periodic reviews of the SRA‟s operations,
performance and effectiveness. Whilst this is a role normally performed by the National
Audit Office, there may be advantages in it being carried out by the Rail Regulator in the
case of the SRA, utilising, the specialist expertise and knowledge of the complexities of the
rail industry available in the ORR.
This new role for the Regulator would provide a clear separation of regulatory and
policymaking roles in the industry. It would provide a proper level of accountability and
redress in the operation of the railway institutions and operators, while removing the
Regulator from direct decision-making and hence removing the source of much conflict and
confusion in the current arrangements.
The incentive framework
The key objective of the proposals put forward in this paper is to enable the railway industry
to operate primarily through commercial incentives in contracts. We have argued that there is
a role for top down decision taking from government to provide the policy framework and
strategy for the industry and this was insufficiently recognised in the original arrangements
for privatisation. It is partly because this government policy framework has been lacking, or
has been inconsistent, that the industry has got into the current mess and the Government has
been drawn into intervening in the industry or exhorting it to do better. In future, the
Government should not intervene in operational decision taking or in contractual matters, and
it should provide more effective support and encouragement to the industry.
Confidence must be restored to those working in the industry to enable it to operate
efficiently. There needs to be a change in the climate of the public debate surrounding the
22
industry and the Government and politicians must to take the lead. The role of the media is
also crucial. Journalists can do much to improve understanding and awareness of the realities
and complexities of running a railway. There will always be scope for criticism of the
inevitable faults, errors of judgement and difficult choices to be made in railway operations
and policymaking. And accidents will continue to happen from time to time on the railway –
at times with tragic consequences - which require urgent and effective investigation.
But the response must be proportionate and rational. The intensity of criticism borne by
railway management and policymakers has undermined morale and has proved very
expensive as the climate of risk aversion has taken hold. There are signs of improvement.
There has been less public criticism of railway performance and more balanced and
informative media reporting of industry affairs of late, while the Government has now
legislated for an independent accident investigation body, apart from the safety and economic
regulators, along the lines successfully used in the aviation industry.
The introduction of a new contractual relationship between the SRA and Network Rail will
provide the opportunity to review all the contracts governing commercial relations within the
industry to ensure that the incentives are properly aligned and that they run back-to-back.
There will inevitably be transitional difficulties.
Franchise agreements now cover the whole gamut of privatisation history – from legacy
longer term contracts such as Virgin Cross-Country, new long-term contracts negotiated by
the SRA, such as Chiltern, and the new long-term contract negotiated by Merseytravel, to
franchise extensions and interim management contracts where more stable arrangements
have not yet been achieved. Aligning all these to the new SRA/Network Rail relationship at
the heart of the industry will be a substantial task; particularly as franchise procurement is
ongoing in parallel.
Similarly, a range of supporting contracts, including rolling stock leases, maintenance
agreements, and station and depot access agreements, will need to be gradually brought into
line with the franchise and track access contracts that drive the bulk of the passenger
industry‟s financial flows and output requirements. The basic framework of most regulated
contracts has already been overhauled by ORR in recent years and the process of fine-tuning
and aligning the commercial terms and incentives should be now be much more
straightforward as a result.
Network Rail will no doubt be reviewing its own supply contracts in the light of their
experience after taking maintenance in-house. Network Rail has also progressed the ongoing
project that it inherited from Railtrack to establish an asset register. It will take time to
populate the register with reliable data and to build up a record of changes in asset condition
and M&R costs over time so as to analyse and forecast M&R requirements reliably. These
initiatives should enable Network Rail to act as an informed buyer and to improve the
effectiveness of out-sourced maintenance in the longer term. It is desirable that a significant
proportion of infrastructure maintenance is outsourced again in the longer term in order to
promote competition and the ability to compare costs and performance between suppliers.
While revenue risk remains with train operators in most cases, they will still require
compensation for unanticipated disruption caused by maintenance, renewal and enhancement
of the network. This compensation requirement has contributed to the infrastructure cost
increases experienced since privatisation, including on the West Coast project. Our proposals
could allow the SRA to specify maintenance, renewal and enhancement requirements of
23
Network Rail under its contract, and to define the complementary franchise specifications,
that minimised the level of compensation payments required.
Major risks
There will remain two major risks to the effective operation of the railway under commercial
incentives with these new arrangements:
the lack of effective incentives for efficiency on Network Rail, and
lack of effective competition in the rolling stock supply market.
Network Rail has a large and diffuse Board and it is unlikely that its governance
arrangements will be effective in holding Network Rail‟s management to account. Its
management disciplines are in many ways weaker than either those of a private sector
company or of a nationalised industry that operates under government scrutiny and
accountability. A management incentive framework has been put in place for Network Rail
but, as with any performance regime, its effectiveness depends on establishing the
appropriate benchmark levels of performance from which to measure under- or over-
achievement. For the reasons described earlier, there is now considerable uncertainty over
the level of efficient costs and performance of Network Rail. But the basic fact, that the
costs of supplying the rail infrastructure have nearly doubled since the accident at Hatfield
indicates that a systematic analysis and investigation into the factors that have caused this
cost escalation is essential, so that costs can be brought back under control. Once this has
been done the management incentive regime can be re-benchmarked appropriately.
Ultimately, Network Rail should be returned to the private sector where it can operate under
private sector disciplines and incentives. However, this is not practical or desirable until it
has established an ability to plan, forecast and control its cost and performance reliably. Until
this is done the risks of infrastructure supply cannot now be passed back to the private sector.
Furthermore, until control is re-established and costs are brought down to a sensible level,
Network Rail‟s performance will remain a major risk to the re-establishment of a viable
railway for Britain.
Since privatisation, the industry has successfully delivered a major programme of rolling
stock investment. However, there is increasing concern that competition in the rolling stock
market is not effective so that the costs of rolling stock leases are too high and the level of
profitability of the three ROSCOs is excessive. A number of factors may contribute to
insufficient competition and high costs in the rolling stock market. These include the limited
supply of existing rolling stock to cater for increased levels of demand, restricted
interchangeability between stock on different lines and the mechanics of the renewal of
leases that are not co-terminus with the end of franchises, so that residual value risks are
priced into new leases. Furthermore, the small size of the railway rolling stock market and
the fragmentation of rolling stock demand between different national railway systems and
„lumpiness‟ in orders, results in high costs and poor efficiency and reliability in the
manufacturing of rolling stock. The fact that rolling stock lease costs are effectively passed
through to the SRA may also limit the incentive on the TOCs to put pressure on these costs.
The Regulator provided a report to the Deputy Prime Minister in 1998 investigating the
competition in the rolling stock market. It found that there were a number of constraints
24
operating which limit the effectiveness of competition but it did not recommend direct
regulation of the market. Rather it suggested that the Regulator should monitor the operation
of the market and use his powers under the Competition Act, if required. Against this
background, it is likely that inefficiency in the structure of the rolling stock market may
continue to add to industry costs in the future. The SRA can play a crucial role in mitigating
this risk by promoting standardisation of rolling stock and „smoothing‟ the ordering of rolling
stock to aid manufacturing efficiency and increase liquidity in the market. It should also take
on responsibility for rolling stock leases over their full life, beyond the end of franchise
terms, to eliminate residual value risk and it should develop a strategy for promoting
competition for new orders between ROSCOs.
There is also a third major risk, and that is that the Government will fail to rise to the
challenge of confronting the hard policy choices that have to made and it continues its ad hoc
intervention, rather than allowing the industry to operate under a framework of incentives.
Conclusions
Seven years on and the Government has still not properly addressed the problems left by
flaws in the Conservative Government‟s privatisation of the railways. „Tough‟ regulation has
proved ineffective as the regulators have presided over an explosion of costs in the industry
and declining train performance, while creeping renationalisation has returned the bill for this
back to the taxpayer. The Government must recognise that the railways need to be run by
commercial incentives and not diktat and its inevitable involvement in the industry must be
structured rationally by developing a coherent policy framework and a streamlined regulatory
structure. It should then step back from direct involvement in the operation of the industry
while providing the political support it needs.
19th March 2004
25
