The Economics of Cost Recovery in Transport: Introduction
Author(s): Kenneth Button
Reviewed work(s):
Source: Journal of Transport Economics and Policy, Vol. 39, No. 3, The Economics of Cost
Recovery in Transport (Sep., 2005), pp. 241-257
Published by: University of Bath and The London School of Economics and Political Science
Stable URL: http://www.jstor.org/stable/20053967 .
Accessed: 03/12/2012 10:54
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
The London School of Economics and Political Science and University of Bath are collaborating with JSTOR
to digitize, preserve and extend access to Journal of Transport Economics and Policy.
http://www.jstor.org
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
Journal ofTransport Economies and Policy, Volume 39, Part 3, September 2005, pp. 241-257
The Economics of Cost Recovery
in Transport
Introduction
Kenneth Button*
There have periodically been sea changes in the way in which the economic
regulation of transport has been viewed. While the traditional divide
between the 'Continental' philosophy of regulation
?
whereby regulation
is seen as the norm to achieve wide socio-political goals and markets only
adopted when this is coincidental with these goals
?
and the 'Anglo-Saxon'
philosophy
?
with its neo-classical economic emphasis on the need to
ensure markets are allowed to work efficiently and unencumbered by
regulators
?
still exists, the focus has shifted very much towards the
latter. We have moved into what Denis Swann and I (Button and Swann,
1989) once called the Age of Deregulation.
Since the deregulation of the US domestic airline industry in 1978, there
has been a global trend towards market liberalisation and privatisation
with the transport sector being particularly affected. The changes that
have occurred have, most economists would seem to agree, been broadly
beneficial. There has been a general downward trend in transport costs
and the use of transport services has risen indicating the revealed prefer
ences of potential users. The US term 'deregulation', although quite
widely used, is, however, something of an exaggeration. There remain
many generic regulations covering such things as mergers policy, labour
conditions, as well as a plethora of safety, security, and environmental
regulations. In addition, in many countries reforms to transport regimes
have been slower in their introduction, and even in those that are consid
*
Professor Kenneth Button is at the School of Public Policy, George Mason University, Fairfax,
VA 20330, USA. Correspondence: kbutton@gmu.edu.
241
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
Journal of Transport Economics and Policy Volume 39, Part 3
ered the most 'free market', regulations, although often inmodified forms,
still remain. What has taken place has been a move to reform regulatory
structures to reduce the most heavy-handed elements of government
intervention.
Where reforms have occurred, while generally beneficial, they have not
always generated the types of results that were expected, and in some case
have been seen as far from successful. Some of these 'surprising' outcomes
are simply the fact that analysts and politicians mis-forecast; a not unex
pected event because a paramount reason for greater reliance on market
mechanisms is that regulators and planners do not have full information
and there are thus often good reasons to leave things to atomised
decision-making structures such as markets. But in other cases foresight
should perhaps have been exercised, at least in general terms.
The aim of this Special Issue is not to look at all facets of the regulatory
changes that have been affected in the transport industries, but rather to
focus on the particulars of cost recovery. The inherent advantage of com
petition is that it provides consumers with choice and forces suppliers to
provide their services as efficiently as possible. A potential problem, long
recognised in the economics literature, is that in some circumstances
there are market imperfections that may lead to inabilities of suppliers to
price in a competitive market in a way that their costs are fully recovered.
In other words not all competitive situations meet the rigid requirements of
perfect competition. In these cases 'adjustments' have to be accepted if
private companies, in our case transport suppliers, are going to provide
optimal levels of service in the long run.
The aim here is not to give short summaries of the papers contained in
the Special Issue
?
hopefully they speak for themselves, and besides there
are abstracts with each
?
but rather to set the scene more fully as to what
the underlying issues are and how over the years transport suppliers
confronted with this cost recovery problem have responded. In this
scene-setting task, I make apology for the fact that all may not accept
some of the points raised. As with a number of areas in economics there
is an inevitable range of interpretations of facts and of the way the world
is seen.
The fixed cost issue
Economic theory tells us that, in open market conditions, when there are no
fixed costs then bargaining between suppliers and customers will ensure
that prices are kept to aminimum consistent with the suppliers recovering
all costs in the long term. Since there are no fixed costs the marginal cost of
meeting the customers' demand will be the entire costs of each unit of
production. Once fixed costs are introduced into the situation and there
242
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
The Economics of Cost Recovery in Transport Button
is competitive supply, then bargaining will push prices down to marginal
cost, but they will be insufficient for capital cost recovery.1
In the transport case, as in many others, this issue was intuitively con
sidered as far back as the mid-19th century when concerns were raised
about the monopoly position of the railroads. These were seen as natural
monopolies but prone to the potential of exploiting this monopoly
power; Harberger allocative efficiency losses, as they are now known,
were feared. The monopoly power itself, however, was seen as needed to
stimulate adequate investment, and without this guarantee of protection
from competition it was felt that entrepreneurs would shy away from the
sector. The result was either legislation that effectively gave monopoly
power to a railroad company over a certain jurisdiction but with rates
of-return regulated so that excess rents could not be earned, or the taking
over of railroads by the state.2
The point here, however, is that the fixed costs were largely seen in
traditional terms of indivisible track, terminals, and marshalling yards,
and this continued well into the 1970s. It subsequently became an issue
again, in a somewhat different context, when it underpinned the unbundling
of the various elements of transport supply that accompanied regulatory
reform in the 1980s. Many measures of privatisation and regulatory
change sought to separate out those elements that had genuine fixed costs
in the traditional sense from those that did not. In the European Union,
for example, elements of airport operations, such as ground handling,
were seen to have small, if any, fixed cost components, and were set free
from public ownership. The provision of road maintenance services in the
UK was similarly treated, and there are many other examples. But that
still left the issue of how to treat the residual elements where there were
perceived to be fixed costs.
Another issue has also emerged. The traditional notion of a fixed cost
may be too limiting for some types of industry or operations therein,
including some within transport. Many transport suppliers offer a sched
uled service entailing a guaranteed product at some pre-specified future
lrTo the best of my knowledge, the Oxford economist Francis Edgeworth (1891) considered this in some
detail as long ago as the 1890s, Jacob Viner (1931) in his seminal work on cost curves raised it, and
subsequently the Nobel Prize winning economist Ronald Coase (1947) developed the topic further.
In practical terms it was certainly an issue to be addressed in the eyes of the economists who were
first asked to review transport markets as part of the European Economic Union's Common Transport
Policy in the early 1960s (European Economic Commission, 1961). The issue of stability is also consid
ered in the context of contestability theory that played some part in the transport regulation debates of
the 1980s (Baumol et al, 1982).
2Canada was something of an exception to this with its fostering of a duopoly system.
3This is the approach found in the main textbooks of the time, e.g. Locklin (1972) and Pegrum (1973).
243
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
Journal of Transport Economics and Policy Volume 39, Part 3
date. This may entail using what are traditionally thought of as fairly
flexible inputs; a scheduled airline flight involves an aircraft that can
always be switched between routes, crew that can be rescheduled, and a
take-off/landing slot combination that can always be used for other
services. In fact, because of the nature of a scheduled service these all
become fixed costs once the timetable is published; the plane is committed,
the crew is committed, and the slots are committed. The fixed cost is thus
not the conventional immobile and inflexible plant and equipment but
rather the commitment of what may traditionally be seen as variable factors
of production to a particular use in the future (Button, 1996).
The challenge confronting the transport supplier is thus how to ensure
sufficient revenue is forthcoming to cover these costs in addition to more
traditional variable factors such as fuel. In the case of scheduled air services
operating in a competitive market, the airline is committed to tying up
capital in each flight, but then has to compete with other carriers for
customers to fill seats. In their quest for business each carrier will quite
rationally reduce prices because providing a passenger covers its marginal
cost it is worth carrying him. This is because a seat is in effect a fixed
cost, perishable product. There is a continual incentive to fill seats that
adds to competitive pressure and short-term excess capacity.
A simple explanation of the underlying problem is offered by Coy
(2002). 'Think about why increasing-returns businesses are unstable. It
costs a lot of money for them to gear up, but once they do, each incremental
unit of output is extremely cheap to produce. Take airlines. Once they fill
their planes enough to cover fixed costs, each additional passenger is
enormously profitable. Trouble is, those profits invariably entice new
entrants. Capacity increases. Fare wars begin. In desperation, airlines
cut fares until some passengers are paying barely more than the cost of
their meals. No longer earning enough to cover their fixed costs,
airlines must merge or go bankrupt. Capacity falls, fares rise, profits
increase, and the cycle begins again.' In the long term when investors lose
confidence in the sector this leads to sub-optimal levels of investment
despite excess capacity often being a short-term problem during peaks in
the cycle.
There are several conditions when there may be an 'empty core', as this
state of the market is known, and, hence, a market may not be sustainable.
It is not only a condition associated with fixed costs, but can also be related
tomatters of avoidable (set-up) costs, indivisibility, or network effects, and
severe fluctuations in demand, where some suppliers enjoy a degree of insti
tutional or financial protection, and when there are significant variations in
the costs of suppliers (Telser 1978). In practice many transport industries
would seem to have cost conditions in which a stable, efficient equilibrium
244
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
The Economics of Cost Recovery in Transport Button
is possible only by means of a suitable restructuring of the competitive
environment.
There is often a tendency to link these sorts of issues with, in particular,
arguments for public subsidies. One way of covering fixed costs, for
example, would be to have them subsidised by the taxpayer. This type of
situation should, though, be distinguished from some of the arguments
for subsidies and market entry controls that have been used to protect
declining transport industries (Kahn, 1988). An argument that has been
widely used in the past is that destructive competition can emerge when
fixed costs are a large part of overall costs and when there are long periods
of excess capacity. In these conditions, the industry is seen as 'unconcen
trated' in that consumers are too few in number relative to the total size
of the market to perceive and to act on the basis of their joint interest to
avoid the competition that drives prices down to marginal cost.
Historically the railways used this line of argument inmany countries
when confronted with lower-cost road competition. Breyer (1984) has
tended to discount the practical importance of this type of problem in
declining industries, seeing it as a short-term structural issue. In the case
of an empty core, the problem may have nothing to do with insufficient
demand but rather the inability of the supplier to structure prices and
services to tap into demand.
The position of the airline industry worldwide has brought a focus more
clearly centred on the cost recovery problem of a growth industry. The com
plexity of the underlying economic model that looks at network industries
offering a committed perishable product has not helped in communication.
The conclusions also run counter to some of the more traditional views of
competition policy where there can never be too much competition.
The analytical framework looking at fixed costs as committed costs
attracted attention in the US in the economic downturn of the early
1990s (Smith, 1995).4 It was rejected at the time by the airlines and by
policy makers who believed that massive cuts in capacity offered the
solution to heavy financial losses being suffered by the scheduled carriers.
The period following this, which saw rapid economic growth in the US,
did see profits being earned, but even at the time of record income for
the industry the hub carriers only managed a net marginal profit of 2.9
per cent. After 2000 losses began to emerge again and have continued to
this day, with numerous carriers being in Chapter 11 bankruptcy. While
there is a need to keep costs down, and to adjust capacity to confront
business cycle effects, which may be difficult, the robust market for the
4
Although for compactness the focus here is on the US airline market similar pictures can be painted of
other transport markets around the world.
245
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
Journal of Transport Economics and Policy Volume 39, Part 3
services of several large low cost carriers over this period in the US does not
seem to indicate chronic excess capacity.5 The issue would seem to be more
intrinsic to the cost structures involved rather than the overall cost levels.
This is not a problem peculiar to airlines and it can be found in other
transport industries; the shipping sector has come in for particular atten
tion. Sjostrom (1989, 1993) examined international liner shipping cartels
(conferences) to seek consistency with an empty core. He looked for legal
restrictions on entry (making the core theory for collusion less likely) or
severe temporal variations in demand and costs (making the core theory
more likely). Looking at the US he found that the theory of the core is
supported. Pirrong's (1992) work put emphasis on exploring costs. He
looked at cost, demand, and market organisation in scheduled shipping.
The findings are consistent with Sjostrom's and suggest that collusion
and coalitions serve to ensure stability and avoid competitive chaos.
The underlying cost recovery problem is far from new and is applicable
to many modern industries beside transport. The financial problems of the
telecommunications sector, for example, seem superficially to be of this ilk.
But here the picture is muddied by its relatively recent deregulation, the
rapid technical changes that are taking place, questionable accounting
practices in some cases, and the nature of the licensing systems in parts
of Europe. Certainly, however, the industry globally is losing money.
Methods of capital cost recovery
Many industries function for considerable lengths of time despite not
recovering their full costs of capital. In some cases this may be possible
because capital debts are written down through institutional means. The
Chapter 11 bankruptcy laws in the US effectively allow a company to be
restructured, and its capital written down, without the entity being
broken up. Several US airlines have made use of this mechanism (some
on more than one occasion). The recent cases of America West, US Air
ways, and United Airlines are examples of this. From a wider perspective,
competing firms often see this as unfair competition because the costs of
their rivals have been reduced through legal means rather than through
strict efficiency improvements.
In other cases industries can attract capital not because the industry as a
whole is seen as viable but rather on the speculation that some companies
5There have been arguments advanced that in network industries, such as air transport, stability can
exist when there are a limited number of suppliers if some can earn relatively high returns on parts
of their networks to cross subsidise other parts while other carriers make their return on other
segments. Mutual interest deters destructive competition with each supplier enjoying a degree of
market power in segments of the system. The advent of point-to-point services can be seen to have
disrupted this, although the evidence to date ismainly North American (Dresner, 2004).
246
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
The Economics of Cost Recovery in Transport Button
within it are. This effort to 'spot the winners' can provide a continual flow
of investment even if the overall probability of making a profit is less than
unity. This 'casino' effect can be heightened when an industry, or some ele
ments of it, has a particular appeal to investors for non-commercial rea
sons. Some sectors simply capture investors' imaginations for periods of
time.
The most common case of this is perhaps sports team companies, but air
transport also has a cachet about it that may attract capital even if it does
not earn a reasonable financial return. In the past the railways were in a
similar position. Most industries are not independent entities but are part
of larger value chains. In some cases those elements in the chain that are
more than recovering their costs of capital may invest in other elements
that are not recovering all costs. It is in their interest to do so either if
they cannot fulfil their role of the loss maker more efficiently themselves,
or if there are institutional constraints against them being direct providers.
Airframe manufacturers often support aircraft purchases with favourable
loans to ensure adequate sales or directly invest in airlines.
Industries also employ a wide variety of techniques to recover their full
cost, or at least more complete recovery. The approach adopted seems to
depend on a combination of the technical characteristics of the industry
concerned, and the institutional environment in which it operates, as well
as the general political acceptability of the system. The following offers a
very brief overview of the main regimes.
Subsidies
Subsidies have long been used to recover the costs of capital and, on
occasions, operating costs. The argument, which goes back to the French
engineering economists of the mid-19th century, is that once an investment
has been made it ismost economically efficient to maximise its use subject
to the willingness of users to pay the incremental costs of their actions. The
ongoing trend is to unbundle attributes of a service and to attempt to isolate
those where the fixed costs are concentrated. These costs can then be
subsidised and the other attributes sold in the market at competitive
prices. This has, for example, been a widely used policy for railways, with
subsidised (and generally publicly owned) track being separated from
commercially driven operations (Brooks and Button, 1995).6
6The UK's rail policy initially involved unbundling with track being placed under a single private
authority while services were tendered for to keep operating subsidies to a minimum. The result has
been a problem for the track authority (where most capital is tied up) to recover costs under a strict
regulatory structure including capital subsidies. The operating companies have found it difficult to
recover their costs even with local market power because of the unreliability of the track network.
247
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
Journal of Transport Economics and Policy Volume 39, Part 3
A problem with any system of direct subsidies is that the incentive struc
ture, unless the subsidies are allocated with care, makes it is difficult to
ensure X-efficiency in production if attained. If the recipient knows that
losses are to be financed from external sources there is no incentive to
resist pay rises for labour or to economise on capital outlays. Further,
there ismuch less of an incentive to provide the goods and products that
customers seek. Tied in with these factors, in some instances there have
been concerns about capture of the system by those seeking economic
rents from the subsidies, either directly or as providers of inputs into the
transport system. Asymmetric information and the nature of the trans
actions costs involved in allocating subsidies, for example, make it difficult
for those giving them to ensure effective use.
A variety of mechanisms have been developed to contain these potential
intervention failures that can accompany direct subsidies. In particular,
tendering for services has become more widely used with 'competition for
the market' being seen as a mechanism to replace competition in the
market. Tendering is not new but it has been finessed in recent years, and
particularly in the UK where, since the mid-1980s, many bus services are
subsidised either on a route basis or, as in London, on a network basis.
While part of the justification for these subsidies is, in different contexts,
to provide social services, and part as a second-best approach to vehicle
congestion policy, there is also an element of fixed cost recovery.
The difficulty even with the more sophisticated methods of tendering is
that while it can potentially introduce more efficiency into the use of subsi
dies, incumbents tend to have an advantage over possible new suppliers; at
the very least they have better knowledge of the market and the tendering
process. There are also challenges in defining exactly what transport suppli
ers are actually tendering for. The funding authority can specify a tight
remit for the tender that allows for a high level of quality control but this
misses one of the key elements of any competitive system, the innovation
that can come from markets. Alternatively a looser set of parameters is
possible but then the authorities inevitably have tomake trade-offs between
the tender and the 'quality' of service that will be provided.
Monopolies
Natural monopolies can recover their full costs by dint of their market
power. In technical terms they have the ability to set either their prices or
their outputs to ensure that revenues exceed costs. The standard analysis
assumes that they establish an output where marginal costs are equated
with marginal revenues and then sell this output at a price that clears the
market. But revenues can be even higher in circumstances where price dis
crimination is possible. At the extreme case, where each potential customer
248
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
The Economics of Cost Recovery in Transport Button
is isolated, the average revenue can be collected
?
hence the airlines' attrac
tion to yield management.
Natural monopolies are, however, by their nature uncommon and are
largely confined to the extractive industries, although some transport infra
structure, such as deep-water ports, do enjoy significant amounts of natural
monopoly. Their rarity in transport is to a large extent because, while a
particular mode may for physical reasons have 'monopoly' over the services
ofthat mode, it services can themselves be substituted by other modes
?
air
travel can be substituted in some cases by rail services, bus services by car,
and so on.
In practice, there is always the tendency for governments to intervene
when there is the potential for a natural monopoly to arise. This may be
through regulations or public ownership. Whatever the case, an inevitable
trade-off emerges between containing the possible inefficiency and exploita
tion that can accompany natural monopoly power, and the potential
intervention costs that can accompany government involvement inmarkets.
But there are also many instances where government consciously creates
monopoly power with the aim of allowing full cost recovery.
Institutional monopoly comes about through such things as licensing
and, where there are subsidies, the tendering out of services. In some
cases the structure is explicit, such as the licensing of a set number of
taxicabs in a city or the tendering out of a bus service, but in other cases
it may be in the form of self-regulation. The latter has been common
amongst the professions engaged in transport
?
engineers, accountants,
lawyers, and so on. This is a subject little studied in the transport economic
literature where labour markets, by and large, are given short shift. Effec
tively, the government gives over the responsibility of limiting the degree of
market entry to the industry itself. This in turn regulates who supplies the
service and at what price. Adjustments can be made to ensure full costs are
recovered. One of the arguments for doing this is that considerable skill and
information are needed to ensure that those in the market are qualified, and
it ismore efficient for professionals to use their resource base to determine
this.
The application of this type of self-regulation can also apply more
directly to transport industries. It has been used, for example, in some air
transport markets for things such as customer's rights, and the Inter
national Air Transport Association has fulfilled this self-regulatory role
in a number of ways over the years.
Institutional monopolies can also emerge as part of wider transport
policy. Licensing systems, for example, often confer monopolies designed
not only to allow full cost recovery on services where there would otherwise
be a willingness of consumers to pay if itwas not for excessive pressures of
249
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
Journal of Transport Economics and Policy Volume 39, Part 3
competition, but also foster cross-subsidisation for services that would not
have been viable even if there were no excessively competitive forces at
work. The bus licensing system operated in the UK from the mid-1950s
had this feature and so did the pre-1978 system of domestic air service
route licensing in the US.
A difficulty with any institutional monopoly, as with a natural mono
poly, is containing the market power that exists. Regulations are widely
used, but their form has changed with time. Traditional rate-of-return
regulation, which was used for many years in the public utilities as well
as transport in the US, allowed prices to recover all costs together with a
reasonable rate-of-return. Beside the inevitable matter of how a 'reason
able' rate of return was to be defined, the limitation of the approach was
that it was frequently captured by the industry under regulation, or by
the regulatory body that had a vested interest in avoiding excessive
confrontation (Stigler, 1971; Posner, 1975). Essentially the industry had
control over cost information and there was no incentive to keep costs
down when any increase could be passed on to customers. Combating
this was difficult for the authorities who, in any case, had minimal incentive
to be too robust in their application of the rules because the appearance of
high complexity justified larger bureaucracies. In some cases workers or
suppliers of inputs further exploited this because they knew there was no
incentive for management to fight input cost increases.
More recent regulatory policies have set price-caps, which have allowed
annual price increases that are at a level below that of changes in the retail
price index
?
the RPI-X formula.7 In the UK this type of regulation is
widespread and applied to telecommunications, energy utilities, and to
water, but it is also used to regulate such transport sectors as the BAA
London airports and the NATS air traffic management system. The
rationale behind the policy is that institutional monopoly is needed to
allow economically efficient provision and full cost recovery but in dynamic
sectors with falling costs there is a need to ensure not only that monopoly
rents are not taken but also that customers benefit from falling costs over
time. Practically, in the UK, while it proved relatively useful in the short
term when formerly nationalised industries had significant levels of X
inefficiency to shed, as costs have moved closer to prices so price-capping
has become more cumbersome and the regimes have moved towards
more traditional rate-of-return regulation. Recent experiences by the
regulator of NATS offers some confirmation of this view.
7Some costs outside the control of the industry (e.g., oil costs) are generally allowed to be passed on
directly.
250
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
The Economics of Cost Recovery in Transport Button
Internal coalitions
Since the inability to recover full costs is frequently due to the excesses of
competition, the suppliers in the market may act to reduce this by forming
horizontal coalitions. At the extreme this may involve efforts to merge or
take over so as to develop, in effect, a degree of quasi-monopoly power.
Much more common in transport are various forms of alliance. These
provide a basis upon which the nature of competition can be adjusted to
allow higher returns to be earned. The creation of shipping conferences
in 1870 was, for example, the first major modern effort to create internal
coalitions within international transport.
Alliances are now common in the airline industry and even extend to
some limited alliances between airports. They appear, at least in their
strategic forms, to have been beneficial to the airlines. The reforms of
freight rail transport in the US have resulted in a number of major alliances
including those across the US/Canadian and US/Mexican borders. The
outcome of these rail alliances has been capacity rationalisation (10,000
miles of track between 1987 and 1999) and a significant growth in produc
tivity and operating income in the 1990s (Transportation Research Board,
2002). A problem, however, is raising funds for investment in new services
while rationalisation is taking place.
One reason for this problem is the nature of traditional antitrust and
mergers legislation in many countries. Alliances often require legal
approval because they have the potential of violating many tenets of com
petition and hence may violate generic antitrust regulations. These regula
tions have traditionally been designed to cope with the potential excesses of
rent seeking monopolists. They have not generally been aimed at con
sidering matters of cost recovery when there may be excessive competitive
pressure in sectors where there are fixed costs. They have even less been
designed to look at cases where these fixed costs are associated with service
commitments rather than physical capital rigidity. In most cases, however,
companies have to raise revenues directly from their sales to recover their
full costs. This is done in practice in a variety of ways but one thing is
common to many of them. The firm needs a degree of market power to
generate sufficient revenue. This type of issue is not always part of
traditional antitrust thinking where ideas of potential excessive market
power form the basis of arguments rather than the desirability of some
degree of market power to allow long-term commercial viability.
Long-term contracts between supplier and customer
By negotiating a long-term cost recovery contract with amajor customer at
the time a service is introduced, a supplier can ensure that there is a
guaranteed revenue flow that will cover most, and, if approached correctly,
251
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
Journal of Transport Economics and Policy Volume 39, Part 3
all, of the capital outlay. This is often standard practice in such traditional
industries as steel production where long-term agreements are reached with
customers. The key features here are that the investments tend to be very
lumpy and relatively long-lived, and that there are not many potential
customers.
Such arrangements are not uncommon in the air transport field. Perhaps
the most discussed is the Civil Reserve Air Force (CRAF) of the US
whereby the federal government gives exclusive contracts to US carriers
for the carriage of their personnel on condition that aircraft and crew are
available at times of national emergency. Similar arrangements have
helped in many European countries. Another example is long-term
contracts between post offices and airlines to carry mail. Such contracts
are also found outside the airline sector, for example between railroads
and power stations for the regular carriage of coal. Branch lines are con
structed on the basis that coal will be carried at a pre-determined rate for
a contracted number of years.
Advanced revenue with subsequent capital adjustments
Most transport activities involve making an investment with fairly predict
able costs and then seek to repay this from much less certain revenue flows.
An alternative is to try to secure a more certain revenue flow and then to
adjust capital outlays so that a viable return may be earned. This is in
effect what many charter airlines do and also many contractual bus opera
tions. They sell capacity in advance to tour operators and others, and are
thus fairly well informed many months ahead of when their capacity has
to be delivered. Freight railway companies in Canada and the US, when
carrying seasonal products such as wheat, pursue similar practices.
As we see in one of the papers in this issue, this approach has been
refined recently in the context of some scheduled airline services. In a
number of US cities groups of businessmen have tried to attract carriers
with guarantees of adequate patronage for an initial period. These are
known as travel banks. InWichita some 400 businesses raised $7.2 million
to attract carriers. Air Tran started operations inMay 2002 with services to
Atlanta and Chicago's Midway airport. It received some $3.0 million to
cover losses in its first year and $1.5 million in the second. Similarly,
Pensacola raised $2.1 million from 319 businesses to attract Air Tran
while companies and individuals in Stockton bought $800,000 of prepaid
tickets to attract American West.
Looked at in a somewhat different way, the full-service airlines engage in
this type of long-term revenue management through frequent flier pro
grammes, and in the past liner conferences have used deferred rebates to
retain regular customer loyalty. These are, amongst other things, intended
252
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
The Economics of Cost Recovery in Transport Button
to reduce fluctuations in the revenue flow and thus inject greater certainty
into longer-term planning.
Vertical integration
If one link in the transport value chain fails to recover its full long-run costs
but the chain in its entirety is viable, then one option is for the loss-making
element to vertically integrate with profitable links. Historically this was
done inmany transport industries when feeder services were vertically inte
grated with the mainline services
?
for example, in the early 20th century
the railway companies often operated bus and trucking services to feed
traffic to their mainline services. It is common practice for many large
companies to self-insure to avoid the costs of outsourcing. In the early
days of aviation airlines were often vertically integrated with aircraft manu
facturers (Boeing and United Airlines being an example), and built their
own terminals at airports, and today the airlines have investments in the
NATS in the UK. In terms of ticket distribution and information systems,
airlines were involved in the development and use of computer reservations
systems. In the US there was direct ownership of the systems until divesti
ture was thought to be a better commercial proposition. The development
of Orbitz in the US and the somewhat less all-embracing Opodo in Europe
reflect a partial return to this form of vertical integration.
Shipping lines have often been involved in developing port facilities;
again to take a UK example P&O heavily invested in the Port of Felix
stowe. Another case where vertical integration has been common has
been the railways
?
although this is changing in Europe. In the US the
Staggers Act of 1980 removed most economic regulations over the freight
railways. The rail companies had control over track and operations. Com
petition pushed down revenues but reductions in capacity (which often
involved selling non-core lines to smaller railroads), reduction in the num
bers of workers, more productive labour practices, and containment over
wage increases meant costs fell even more. In 1986 the railroad operating
expenses were 93 per cent of costs but by 1995 they were 86 per cent of
costs. Basically, the railroads had control over a major input cost
?
their infrastructure.
The difficulty with vertical integration is that it imposes additional man
agerial strains to the system. It essentially often moves transport service
suppliers away from their core business and reduces the efficiency with
which they can determine their profit centres. What it also often does is
to move economic rents down the value chain, transferring them from
elements where there ismonopoly power to those where high levels of com
petition make the recovery of fixed costs more problematic. While this is a
potential second-best approach to full cost recovery, the overall adverse
253
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
Journal of Transport Economics and Policy Volume 39, Part 3
effects on efficiency may be very high and the outcome far removed from a
first-best world.
Discriminate pricing
Discriminate pricing is widely practised in transport and other network
industries. The underlying idea was summarised a few years ago by the
UK Office of Fair Trading (1999). 'In general undertakings will need to
set prices above their incremental costs so that common costs, for example,
can be recovered. Price discrimination between different customer groups
can be a means of achieving this; it can increase output and lead to
customers who might otherwise be priced out of the market being served.
In particular, in industries with high fixed or common costs or low marginal
costs, it may be more efficient to set higher prices to consumers with a
higher willingness to pay.'
Price discrimination has been widely used in transport to recover costs.
It was used almost at the outset by shipping conferences (Goss, 2002) and
continues today, albeit in a modified form, even as maritime transport
moves into the container age (Brooks and Button, 1996). It was developed
and refined in dynamic form in the US domestic air transport market as
yield management during the 1980s; essentially fares change as a plane
fills. Peak and off-peak fare differentials are common for local public trans
port systems. Regarding 'information transport', deregulated long-distance
telecommunications providers give quantity discounts to both large and
small customers; charge business and individuals different rates; and offer
calling plans that offer discounted rates based on individual characteristics
and usage patterns.
The idea of yield management is to extract as much revenue from custo
mers as possible by levying prices that reflect the willingness of customers to
pay. Consequently, customers less sensitive to price pay more and contri
bute to the capital cost of the service, while those who are less willing to
pay are charged lower prices that at least cover their marginal costs. It
has been long used in a particular form to recover the capital costs of
various public utilities such as passenger rail services and urban bus
services. Here the aim is not to price differentiate between users so as to
maximise profit but rather to generate revenue so that an acceptable
return is earned after all costs (including the cost of capital) have been
covered. This approach, known as Ramsey pricing, constrains the level
of prices levied at the upper end.
Two-part tariffs
Explicitly separating out capital from operating or marginal costs and char
ging for each separately has been a standard way for many utilities to
254
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
The Economics of Cost Recovery in Transport Button
recover their full costs. In the case of telecommunications (be it hard wire or
cellular) consumers normally pay an access fee (roughly equal to the capital
cost) and a usage fee. In some cases, if the marginal costs are extremely low,
there is simply an access fee that subsumes the estimated costs of individual
calls. This is a standard practice for local telephone services. The two-part
tariff is also to be found in a variety of private sector activities
?
for
example, the golf club membership and the green fees.
The successful use of two-part tariffs depends very much on both the
relative importance of the capital element in costs and the frequency with
which individual consumers use the good or service. It is generally far
more successful when there is regular and heavy use. This is because the
supplier can employ information to adjust pricing and investment plans
and consumers do not feel a heavy burden when it comes to paying the
periodic 'membership' fee. It is also most efficient when the service being
offered is fairly homogeneous. This makes allocating the fixed fee easier.
The use of strict two-part tariffs in transport is relatively limited because
most people do not travel often enough tomake it an attractive option and,
from the transport operators' perspective, the nature of trip making means
that it is difficult to allocate a reasonable approximation of capital costs to
each consumer. Some road charging regimes may, however, be seen as
approximating to a two-part tariff (an annual licence fee plus a variety of
user charges). The link between costs and prices are, however, rather
opaque even in regimes where there are road funds that dedicate at least
part of the taxes and charges road users pay to infrastructure use and
operation (Newbery, 1988). One problem is that in many countries at
least part of the payments made by road users are seen as sumptuary
taxes unrelated to cost recovery.
Some final thoughts
Transport economics is essentially concerned with problem solving within
particular institutional and technical constraints. The issue addressed in
various ways in this Special Issue is how the challenges of full cost recovery
can be achieved in transport markets. The very nature of the papers clearly
reflects Coase's point that there is no single way of recovering full costs in
competitive situations where there are any forms of fixed costs. There are
multiple ways, some more efficient than others, but efficiency is not
always easily discerned in situations where information is not complete,
and in many cases asymmetric, where there are obscure transactions
In a way this may be seen as form of price discrimination in that average costs of trips vary according to
the number of trips made and thus ultimately reflects willingness to pay.
255
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
Journal of Transport Economics and Policy Volume 39, Part 3
costs, and where the motivations of the actors do not conform to the ideas
of public interest theory. The challenges are compounded when political
economy is enmeshed in neo-classical positivism. In reality, in many
cases itmatters as much who pays for the full costs of transport as how
the efficiency of the system is achieved.
References
Baumol, W., E. E. Bailey, and R. Willig (1977): 'Weak Invisible Hand Theorems on the
Sustainability of Multiproduct Natural Monopolies,' American Economics Review, 67,
350-65.
Bittlingmayer, G. (1982): 'Decreasing Average Cost and Competition,' Journal of Law and
Economics, 25, 201-9.
Bittlingmayer, G. (1985): 'DidAntitrust Policy Cause theGreat Merger Wave?,' Journal of
Law and Economics, 28, 77-118.
Breyer, S. (1984): Regulation and its Reform, Harvard University Press, Cambridge, MA.
Brooks, M. and K. J. Button (1995): 'Separating Track from Operations: a Look at
International Experiences,' International Journal of Transport Economics, 22, 235-60.
Brooks, M. and K. J. Button (1996): 'The Determinants of Shipping Rates: a North
Atlantic Case Study,' Transport Logistics, 1, 21-30.
Button, K. J. (1996): 'Liberalising European Aviation: is there an Empty Core Problem?,'
Journal of Transport Economics and Policy, 30, 275-91.
Button, K. J. and D. Swann (eds) (1989): The Age of Regulatory Reform, Oxford Univer
sity Press, Oxford.
Coase, R. (1946): 'The Marginal Cost Controversy,' Econ?mica, 13, 169-89.
Coy, P. (2002): 'Deregulation: Innovation vs. Stability,' Business Week, 28 January, 108-9.
Edgeworth, F. Y. (1881): Mathematical Physics, Kegan Paul, London.
European Economic Commission (1961): Memorandum on the General Lines of a Common
Transport Policy, EEC, Brussels.
Goss, R. O. (2002): 'An Early History of Maritime Economics,' International Journal of
Maritime Economics, 4, 390-404.
Kahn, A. E. (1988): The Economics of Regulation: Principles and Institutions, MIT Press,
Cambridge, MA.
Leibenstein, H. (1966): 'Allocative Efficiency, X-efficiency and the Measurement of
Welfare Losses,' American Economic Review, 56, 392^15.
Littlechild, S. C. (1983): Regulation of British Telecommunications' Profitability, London,
Department of Industry.
Locklin, D. P. (1972): Economics of Transportation (7th ed.) Richard D. Irwin, Homewood.
Newbery, D. M. G. (1988): 'Road User Charges in Britain,' Economic Journal Supplement,
98, 161-76.
Pegrum, D. F. (1973): Transportation Economics and Public Policy (3rd ed.), Richard D.
Irwin, Homewood.
Pirrong, S. C. (1992): 'AnApplication of Core Theory to theAnalysis of Ocean Shipping
Markets,' Journal of Law and Economics, 35, 89-131.
Posner, R. A. (1975): 'The Welfare Effects of Monopoly and Regulation,' Journal of
Political Economy, 83, 807-27.
256
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions
The Economics of Cost Recovery in Transport Button
Sjostrom, W. (1989): 'Collusion inOcean Shipping: a Test of Monopoly and Empty Core
Models,' Journal of Political Economy, 97, 1160-79.
Sjostrom, W. (1993): 'Antitrust Immunity for Shipping Conferences: an Empty Core
Approach,' Antitrust Bulletin, 38, 419-23.
Stigler, G. (1971): 'The Theory of Economic Regulation,' Bell Journal of Economics, 2,
3-21.
Telser, L. G. (1978): Economic Theory and the Core, University of Chicago Press, Chicago.
Telser, L. G. (1985): 'Cooperation, Competition, and Efficiency,' Journal of Law and
Economics, 28, 271-95.
Transportation Research Board (2002): Freight Capacity for the 21st Century, TRB,
Washington DC.
Viner, J. (1931): 'Cost Curves and Supply Curves,' Zeitschrift f?r National?konomie, 3,
23-46.
UK Office of Fair Trading (1999): Assessment of Individual Agreements and Conduct, OFT,
London.
257
This content downloaded by the authorized user from 192.168.72.232 on Mon, 3 Dec 2012 10:54:19 AM
All use subject to JSTOR Terms and Conditions