Competition and Ownership in Land Passenger Transport:
The Sixth International Conference
Department of Transport Economics, Rand Afrikaans University, Johannesburg, South
The aim of this paper is to provide the background to the Sixth International Conference on Competition and Ownership in Land Passenger Transport, held in Cape Town, South Africa, that was attended by 161 delegates from 23 countries in September 1999. We review past conferences in the series and summarise the structure of the Cape Town conference and the key emerging issues. Further details are given by the five workshop reports that accompany this introduction.
1. Review of Past Conferences
Details of the previous five conferences are summarised in Table 1
(Preston and Nash 1998). The first conference in this series was instigated by Professors
Michael Beesley and David Hensher and was held in Thredbo, New South Wales, Australia in May 1989. The key papers were published in a special edition of Transportation Planning and Technology in 1991 (Hensher et al 1991). This conference focused on bus and coach services. The policy background was dominated by the research emerging from the deregulation of local bus services in Great Britain outside London in 1986 (competition-in-the-market) and the experiments with competitive tendering (competition-for-the-market) in London and a number of US cities. Some of the major themes that emerged from the conference were:
1. The need to break the nexus between services and subsidy.
2. The assessment of cost savings associated with different forms of market arbitration. 3. The role of the minibus as a major technological enhancement.
4. The role of competitive contracting and the need for an open book approach. 5. The importance of travel cards and other intermodal ticketing.
6. The role of performance and productivity measurement.
The second conference in this series was at Tampere, Finland in June 1991 with the
proceedings published by Viatek (Beesley et al 1992). The euphemism 'competition and ownership' in the conference title was replaced by 'deregulation and privatisation', but a more significant, and longer lasting, change was that 'bus and coach' was replaced by 'passenger transport'. Rail systems were considered for the first time, as was road pricing and investment, whilst the conference also had a temporary dalliance with airlines. Competition policy (anti-trust to Americans) emerged as an issue and had its own workshop, whilst externalities were considered - not just congestion, but also safety, use and non-use values, the environment and
;All conference papers are available for downloading at the Institute of Transport Studies website: http://www.its.usyd.edu.au/thredbo_papers.html. One can also join the Thredbo discussion
list from this site.
impacts of urban form and life styles. Of the seven themes that summed up the second conference, two and a half were continuations of the first conference's themes. The discussions on the role of market and alternative forms of intervention and on the progress in contracting and tendering were a continuation, as was the discussion of new approaches to encourage competitive efficiency, at least in terms of the functional separation of intermodal ticketing. Other new approaches discussed included property development and value capture, allocations of land / zoning and the privatisation of planning. The other four themes were new to the conference:
1. The need to determine the proper role for pro-competition agencies.
2. The need to establish a framework for modal rivalry.
3. The need to promote the political acceptability of reform agendas.
4. The need to establish better theories of technique and planning practice.
The third conference in the series was held in Toronto, Canada in September 1993, with the proceedings published by the Ontario Motor Coach Association (Love, 1994). The conference's coverage of surface passenger transport embraced for the first time taxi, light rail and guided bus. The conference's familiar themes of the impact of competition and competition policy and of competitive tendering continued to be explored. An important contribution was provided by the workshop on infrastructure policy for roads and railways where detailed recommendations were made concerning the stages in implementing an infrastructure program and concerning the maintenance of, operation of and access to infrastructure. The role of economic regulation, must notably price capping and licence conditions, was also assessed.
The fourth conference was held in July 1995 at Rotorua, New Zealand. The full proceedings were produced by Transit New Zealand (1995), whilst summaries of the conference and its workshop reports were published in Transport Reviews (Hensher and Knight 1996).The policy backdrop for the conference was the liberalisation of the New Zealand economy. This resulted in the Transport Law Reform Act of 1989 which (in 1991) deregulated bus and coach services and the privatisation of New Zealand Rail in 1993. Three of the five workshops examined the now familiar themes of competitive models and impacts; regulatory reform and transport policy development; and international experiences in competitive operations. The rail sector had its own workshop as did user requirements.
The fourth conference detected an overemphasis on cost-minimisation ('doing the thing right') at the expense of user requirements ('doing the right thing'). Improved understanding of the relationships between contract duration and incentives appeared to offer a prospect of being able 'to do the right thing right'. Other themes that emerged were the need to ensure quality and transparency of data (assessment of the New Zealand reforms appeared to be hampered by lack of data), to determine efficient transport prices (and in particular determine the role for congestion pricing and complementary non pricing actions) and to determine the appropriate relationship between infrastructure owners and operators (and especially access and pricing mechanisms).
The fifth conference was organised against the policy backdrop of the election of the new Labour Government in Britain with commitments to increasing regulation of the bus and rail sectors, and the European Commission's Green Papers on the Citizen's Network and Fair and Efficient Pricing. The principal interest remained in land-based public transport, particularly bus and rail transport, but with interest also in the interaction between public transport and private road-based passenger transport.
One of the conclusions arrived at the conference was that competitive forms adopted by
different countries will depend on and be tailored to national or regional administrative or entrepreneurial capabilities, as well as to initial market conditions. Discussions were also held about the various forms of competition for-the-market by making a distinction between strategic, tactical and operational functions. No consensus could be reached on an optimal form of contract specifications. Discussions also centered on quality partnerships as a practical means of low level regulation as well as incentives that are considered necessary to achieve outcomes that are consistent with the broader goals of transport policy. The workshop reports were published in Transport Reviews in 1998 (see Preston and Nash 1998).
2. The structure and Key Themes of the Cape Town Conference (Thredbo 6)
The aim of the Cape Town Conference was to provide an international platform to discuss passenger transport competition and ownership issues. As South Africa is in the process of implementing a new policy on competition and ownership in public transport, and has advanced far in this process in terms of especially bus transport, many of the discussions centered on international experiences and the lessons South Africa could learn from such experiences. The conference was therefore well attended by National, Provincial and Local Government delegates.
The conference was organised along a series of plenary papers on the morning of the first day thof the conference which reviewed policy changes in the 20 century and new policy directions stfor the 21 century. The conference also had five parallel workshops, the details that are summarised in Table 2. A list of objectives were given to the session chairmen as broad guidelines to assist them in managing their respective workshops.
Workshop 1 focussed on competition in public transport with the aim to provide an assessment of developments and trends in tendering and competition in public transport. The papers discussed different organisational models for land based passenger transport. These organisational models could be grouped into three broad models: The first model is that of free competition with residual regulation - the British model. The second model is that of public ownership and regulation and was not discussed explicitly. The third model was a set of models based on competitive tendering . Three variants were identified of the third model; firstly the Scandinavian model pertaining to ferries in Norway, bus and rail systems in Sweden, and less obviously, to bus in certain cities in the United States. Secondly, the French model of management contracts and thirdly hybrid models in which quality as well as price was taken into account. The latter models are being considered by the European Commission, Italy and South Africa.
Workshop 2 on funding of public transport and infrastructure differentiated between funding and financing to avoid ambiguities. The workshop considered whether and how government funding might be associated with pre-determined outcomes as well as the sourcing of government and other funding. A number of inter-country comparisons were drawn. A number of papers considered the specific contexts of transport agencies in Canada, France, South Africa, Sweden, New Zealand and Britain. The funding roles of various levels of government were also discussed.
Workshop 3 on user needs and impact on public transport focussed on the theme of quality in public transport. The workshop considered how international experiences of performance of public transport (and its monitoring) could be used to the benefit of specific countries. The complexity of informal transport service delivery and risks in formalising it to such an extent that its contribution is minimised, if not eliminated, were discussed. The workshop made eight recommendations regarding aspects such as user needs, quality partnerships, diversity of basic
needs, development of guidelines on how service quality should be measured etc.
Workshop 4 addressed key areas associated with the management of public transport systems. The key themes that emerged from the resource papers dealt with the regulatory reform process, the institutional response to the regulatory reform process and performance evaluation in terms of the public transport system response to these processes. The workshop concluded by stating that public transport is operating in an environment of constant change and that transport management has, out of necessity, to be on the boundary of continual and dynamic change - challenging the existing and entrenched views and the stimulation to move along the constructive spiral of change.
Workshop 5 on ownership and organisation of public transport and infrastructure dealt with, amongst others, the developing and transitional economies, "economic empowerment" in the transport reform processes, the relationship between quality and regulatory reforms as well as rail reform initiatives in Western Europe, South Africa and China. The workshop concluded with key findings and a number of recommendations.
Public transport is in a phase of continuous and accelerated change throughout the world. Although numerous approaches are followed in countries around the world the central theme of change is to achieve higher levels of efficiency and effectiveness in transport systems, and where applicable, to stretch the value for money of public subsidies. The emphasis on competition to achieve these objectives is evident, through the tendering for contract system, management contracts, privatisation of state assets and public - private partnerships, performance benchmarking etc. These developments are important for public transport practitioners. It is also necessary to reflect on the effectiveness of the new measures and share the experiences with each other. The THREDBO series of conferences serve as the ideal vehicle to achieve this as it is focussed and entirely relevant in today's ever-changing transport environment.
Beesley, M., Hensher, D. and Talvitie, A. (1992) „A Summing Up: Privatisation and Deregulation of Passenger Transport‟. In „Privatisation and Deregulation in Passenger Transportation. Selected Proceedings of the Second International Conference‟. Viatek, Espoo, Finland, pp5-12.
Hensher, D.A., Battellino, H. and Beesley, M.E. (1991) „Competition and Ownership of Public Transit Services: Introduction‟. Transportation Planning and Technology, 15, 2/4, 85-94.
Hensher, D.A. and Knight, F. (1996) „The Fourth International Conference on Competition and Ownership in Land Passenger Transport‟. Transport Reviews, 16, 3, 183-212 (Part 1) and 16,
4, 277-299 (Part 2).
Love, J. (Ed.) (1994) „Proceedings of the Third International Conference on Competition and
Ownership in Surface Passenger Transport‟. Ontario Motor Coach Association, Toronto, Ontario.
Preston J. and Nash C. (1998) Competition and Ownership in Land Passenger Transport: The Fifth International Conference, Transport Reviews, 18 (4), 321-327.
Transit New Zealand (1995) „Fourth International Conference on Competition and Ownership in
Land Passenger Transport: Conference Papers‟. Transit New Zealand, Wellington.
Tendering and Competition In Public Transport: Theme 1
JOHN PRESTON and ANDREW SHAW
The aim of this workshop was to provide an assessment of developments and trends in tendering and competition in public transport. The workshop was based around 16 resource papers (listed in the Appendix) and discussion involving the 56 workshop delegates. The format of this report is as follows. In the next section, the evidence provided by 16 resource papers will be briefly described and some common themes identified. In section three, some key points of discussion will be highlighted. In section four, some tentative conclusions will be drawn. Policy context was provided by the on-going reforms of public transport markets in South Africa.
2. The Evidence
Staffan Hultén‟s first presentation described three rounds of rail franchising in Sweden. In the first round, a small entrant B K Tåg made some gains. In the second round, these tenders were recaptured by the state railway, SJ. In the third and current round, a number of operators (including B K Tåg) have made gains from SJ. The dynamics of tendering are clearly important - and not always easy to predict.
Wendell Cox examined evidence collected by Martin Wachs and colleagues (McCullough, Taylor and Wachs, 1998) that purported to show that contracted public transport services in the US were more expensive than non contracted services. Wendell comprehensively destroyed this assertion and showed instead that competitive tendering had led to cost reduction of almost 50%.
Doug Crockford illustrated the huge variety in competition and private ownership in the European Union but showed that this did not correlate with the large variations in financial support. For example, in Ireland little financial support is required despite the fact that buses are, in the main, regulated and publicly owned.
Ken Gwilliam showed that tendering in Uzbeckistan was initially a failure due to a long list of qualitative requirements, including donations to local charities. However, with sufficient will, the tendering system has subsequently been made to work. We clearly need to be careful about qualitative criteria.
Olav Hauge showed that the tendering of ferries in Norway also seemed to be successful, although there is an important issue of safety regulation that needs to be considered (particularly given the subsequent accident on the Stavanger-Bergen route).
From Staffan Hultén‟s second presentation we learnt that, in Sweden, the State Railway, SJ, could not resist opening the Pandora‟s box of competition. This paper reminded us of the
dangers of regulatory capture, although in this case, thankfully, the state monopolist seems to have failed in its attempt to capture policy makers. In fact the reverse seemed to occur.
Kjell Johansen presented a technical paper on bus costs in Norway which showed that some forms of competitive pressure, based on normalised cost contracts, which fall short of tendering,
can lead to modest additional cost reductions. However, there was little evidence of economies of scale in Norway, other than the biggest operator had 10% lower unit costs.
Ferdinando Stanta looked at simulating bus tendering in Lombardy, Northern Italy. Operations research techniques were used to determine the areal extent of tenders (around 50 were specified for an area with a population of 9 million), the Public Service Obligation (especially in terms of capacity); the financial resources and the evaluation of (artificial) bids. Multi-criteria analysis was used so that a number of criteria in addition to price might be assessed.
Rosario Macario was also concerned with ensuring that quality as well as price was considered when considering competition models in the European Union. However, is this a legitimate attempt to produce a more sophisticated competitive regime or an attempt to forestall competitive tendering? A second question is can contracting out of planning as well as operations be feasible?
Chantelle Barbieux told of the failure of deregulation in the UK, Chile, Algeria and Morocco. The French alternative was a competitive model but framed with contractualisation and regulation.
Enilson Santos discussed the role of combi-vans in Rio de Janeiro, Brazil, which have high customer satisfaction but a relatively small market share (5%). In South Africa where combi-vans have a market share some ten times higher, do these services have a similar high customer rating?
Lena Nordenlöw discussed how small firms could stand tall in the shadow of giants. One solution is for small firms to join alliances and associations but this can fall foul of the anti-trust agencies, as has occurred recently in Britain with quality bus partnerships.
Baard Norheim discussed ways of reducing costs (by 20%) and increasing demand (also by 20%) in Oslo through performance contracts rather than competitive tendering. However, there is a need to assess the counter factual - how can we be sure that competitive tenders could not have done better?
Øyvind Sunde used game theory to explain how firms might adopt strategies to manipulate tendering experiments. He concluded that we have to be careful when interpreting the results of one-off demonstration projects.
Pascal Vincent described the franchising process in France. In particular, the Sapin law was introduced in 1993 to ensure fair competition in tendering, following a series of corruption scandals. It is a three stage procedure (candidature, bidding, negotiations) that takes on average over 8 months - suggesting relatively high compliance costs.
Lastly, Jackie Walters described the scope for franchising and sub-contracting in South Africa and in particular examined the ways that small and medium enterprises might be involved in this competitive process. There was seen to be some scope for McDonald‟s style franchising,
but not for British rail style franchising.
The unifying theme of the papers was they were all examining different organisational models for land based passenger transport (the ferries considered in Norway predominantly carried cars and could be thought of as an extension of land based modes). These models fall into three broad types (but see Van de Velde, 1999, for an alternative classification). The first model, that of free competition with residual regulation - the British model, was not covered
directly but was often postulated as a bench mark. Brazil and South Africa provided de facto examples. Similarly the second model, that of public ownership and regulation was not explicitly covered other than as a bench-mark, although the paper on Oslo suggested an interesting variant involving what might be called residual competition. The focus of the workshop was on the third set of models based around competitive tendering. At least three variants were identified. First, the Scandinavian model, which naturally enough pertained to ferries in Norway, bus and rail in Sweden, and, less obviously, to bus in certain cities in the United States. Secondly, the French model of management contracts was discussed by two papers and may be the way British rail franchising is evolving. Thirdly, hybrid models in which quality as well as price was taken into account. Such organisational models were being considered by the European Commission, Italy and South Africa and a variant had been implemented in Uzbeckistan.
3. The Discussion
Initial discussion revolved around four areas. First, the degree of restructuring that is required to ensure that there is fair competition for contracts - the level playing field. The following issue were identified:
1. The need for the functional separation of operators and authorities. The contracting
agency needs to be seen to be even handed. This may not be possible when there are
arm‟s length relationships between authorities and publicly owned operators.
2. Competitive and non-competitive activities need to be ring fenced to prevent contract
bids being cross subsidised.
3. Publicly owned operators may need to be privatised before tendering can be successful. 4. Large firms may have advantages when it comes to tendering and therefore there may
be a need to encourage entry by small firms. However, it was widely believed that the
bus industry exhibited constant returns to scale and although large firms may be able
to obtain vehicles, fuel and capital on favourable terms, small firms often face lower
labour costs thus offsetting these advantages.
Secondly, the specification of the contract was discussed. The merits of net cost and gross cost contracts were examined. More consideration is required of hybrids such as output based contracts. In terms of size, the merits of area and route based contracts were discussed, with some presumption in favour of the latter. The merits of operating contracts (in which authorities provide the vehicles) and complete contracts (in which the operators provide the vehicles) have been considered particularly with respect to recapitalisation requirements. Length has also been considered, particularly with respect to Sweden where contracts as short as one year have been let.
Thirdly, with respect to contract evaluation, an important but difficult trade-off between price and quality was identified.
Fourthly, there was discussion about contract enforcement and in particular how to devise incentives to ensure self-enforcement. This revolves around the identification of an appropriate system of penalties and bonuses.
Two generic questions were also posed. First, is it possible to get it right first time so that competitive tendering leads to both subsidy reductions and quality improvements or is it inevitable that subsidy reductions must come first? Second, are different organisational models appropriate for different modes (minibus, taxis, bus, rail, ferries)?
The conclusions of the workshop were an extension of the discussion and revolved around the relationships between the Public Transport environment and contract competition, contract specification, contract award, contract enforcement and contract renewal (see Figure 1).
Insert Figure 1 near here
Ten tentative conclusions may be drawn:
1. Different organisational models may be appropriate in different circumstances. Free
market models may be most appropriate for bus and minibus in small towns and niche
markets. Some form of Scandinavian competitive tendering may be appropriate for
buses in larger towns and for ferries. What we refer to as franchises or concessions
may be more appropriate for rail. The appropriateness of different models may also
vary by the time and place. Hybrid models are likely to appeal to continental Europe
with its long tradition of the classic regulation model. Temporary freeing up of the
market may be useful to determine the degree of productive and dynamic inefficiency
believed to be inherent in regulated markets, but may be followed by some form of re-
2. It may not be necessary to experiment directly in the market place. Ferdinando
Stanto‟s paper showed that simulation models might be helpful in determining
appropriate contracts. Preston et al. (2000) have come to similar conclusions with
respect to rail. By contrast, Øyvind Sunde was sceptical of the worth of demonstration
projects. Further evidence for this might be the mixed and at times misleading
messages that came from the bus deregulation trial areas in Britain in the early 1980s
(Fairhead and Balcombe, 1984, Evans, 1988).
3. Even when competition is increased in the public transport sector, residual regulation
may also need to be increased. Obvious examples include safety related measures (e.g.
operator, driver and vehicle licensing) and anti-trust. The workshop also highlighted
labour protection. One concern about competitive tendering is that it can become an
auction in the labour market. Where there is no minimum wage legislation nor a
comprehensive welfare system (as is the case in most developing countries) there may
be danger that wage rates are driven too low. This is a possible example of the well
known winner‟s curse phenomenon in auction theory.
4. Measures need to be taken to ensure that tendering is competitive in initial and
subsequent rounds. Recent evidence from London indicates what a difficult task this
may prove to be (Beaumont and Watson, 1999). The workshop agreed that
institutional reform would be needed, in particular to separate the planning
responsibilities of government from service delivery responsibilities and to ensure a
level playing field. Organisational reforms would probably also be needed. As a
minimum, an active private sector competitive fringe is required. It seems sensible to
unbundle large state owned transport enterprises and to commercialise their
management, possibly in preparation for privatisation. Anti-trust legislation needs to
be alert to the potential for re-agglomeration to lead to local monopolies or cartels,
although the evidence that large firms have advantages in competitive tendering is
relatively weak, at least for route based contracts. Policy should be pro-active in
encouraging the participation of small and medium enterprises and, for developing
countries, engaging the so-called informal sector with the formal sector. 5. It was agreed that contract specification was of critical importance. In the workshop
we heard about contract lengths of between 6 months and 30 years. The consensus
seemed to be that contract lengths of 4 to 5 years were appropriate for the bus industry,
but that contracts should be longer for rail, particularly if operators are responsible for
providing rolling stock. Operating contracts in which authorities and/or third parties
provide the vehicles/rolling stock to operators may be a useful device to inject more
competition into the market, but as the market matures complete contracts might also
be considered, particularly for bus but also for rail. There was no consensus on the
relative merits of gross cost and net cost contracts despite some empirical support for
the former (White and Tough, 1995). However, the lack of consensus may become
increasingly irrelevant given the move towards hybrid contracts based on gross cost
plus revenue incentives (so called output based funding). An important issue with
respect to contracts that was unresolved was their areal extent. Route-based and area-
wide contracts both had their supporters, although it is probably fair to say that too
small contracts are less of a problem than too large contracts because they can be re-
packaged by bidders into larger contracts. The workshop was also concerned that
contract specification ensured integration within and between public transport modes.
Similarly, there was concern that contract designs were too top down and a bottom-up
approach in which users were more widely consulted might be preferable. 6. It may be possible to consider contracting out the planning function as well as the
operating function. Evidence from cities that have made some progress in this respect,
such as Helsingborg (Sweden) and Indianapolis (U.S.A.), needs to be collated. 7. There was some concern that evaluating contracts with different levels of quality might
be problematic. However, this workshop concurred with workshop 3 in the belief that
such comparisons were possible using service quality indices and similar tools
developed by projects such as QUATTRO (European Commission, 1998).
8. There was a consensus that the contract award stage should be transparent, with as
much information as possible in the public domain.
9. Enforcement costs for regulatory bodies were believed to be substantial, with detection
of contract breaches particularly difficult. Contract withdrawal seemed to be the most
effective deterrent. Although it was believed that financial penalties have a role to play,
they need to be at relatively high levels (and accompanied with adverse publicity
whenever these penalties are incurred) in order to provide sufficient incentives. 10. Contract renewals need to be dealt with carefully. Contracting authorities should be
prepared for contractors to go bust. They should have contingency plans for such
eventualities to ensure continuous service delivery that do not involve baling out the
original contractor. There was some suspicion about mid term contract re-negotiations
and automatic extensions, which might in the long run trigger the need for Sapin style
One workshop member emphasised the importance of the 7Ds: decentralisation; distintegration; domain of the contract; discretion of management; distribution of risk; duration of the contract; and destination of the subsidies. The workshop‟s most important contribution was probably to thdraw attention to an 8 D: the dynamics of tendering and competition in public transport. The workshop believed that in many cases the public transport environment will need to be reformed before tendering or competition can be introduced successfully. Even if the first round of tenders is competitive, attention needs to be paid to ensuring this is the case in second and subsequent rounds. Contract specifications will need to be continuously updated to take into account the changing balance between cost reductions and quality improvements and changes in the external environment. Getting it right now doesn‟t guarantee it will be right in the future.