Bus contracts throughout the world have undergone gargantuan changes over the last 40 years as many countries have moved away from nationalised and publicly owned bus services. While there have been many benefits from this move, in recognition of the advantages of opening up the local bus service market to competition, the pendulum has swung away from a fully economic deregulated model towards one where competition for the market recognises the natural monopoly characteristics of local bus service delivery.

We now see a mixture of competitively tendered (CT) contracts and negotiated performance-based contracts (NPBC) in many geographical jurisdictions, with the choice commonly driven by a view that CT ensures transparency and best value for the taxpayers’ dollar (given that such services are essentially funded from government budgets), whereas NPBCs lack such transparency. This claim has eroded over the years as we have come to see evidence that cost efficiency and network effectiveness outcomes are often as good as, if not better, under NPBCs compared to CT.  Competitive tendering will not solve everything; NPBC’s also have value, but they come with conditions (as do and should CTs), and benchmarking is essential in both cases.

Hensher and Stanley (2010) reviewed the evidence together with numerous studies reported in the Thredbo series, suggesting that the case for tendering is no guarantee of achieving a cost efficient (‘value for money’) and service efficient outcome that is a significant gain over a well-designed negotiated contract model accompanied by a well-articulated performance regime with monitored key performance indicators. Despite the calls for careful comparison, competitive tendering in the main has won out on the claimed argument of transparency. Hensher (2015) questioned this and provided evidence that many negotiated metropolitan contracts in Australia and elsewhere have been more cost efficient and hence give greater value for money than tendered contracts. The argument that competitive tendering is the best way to refresh the market has not been proven to be as effective as incumbent operators selling out to mainly multinational bus operators in many jurisdictions globally. Indeed, the greatest quantum of businesses changing in many jurisdictions in Europe and Australasia, in particular, has occurred through such purchase and not tendering. Using data linking CT prices of successful bids to NPBC outcomes (Hensher 2015), the evidence suggests that the gains from CT are often illusory (outside of the situation of an incumbent public operator). These findings send a strong message about the presumption that competitive tendering is necessarily the only way forward.  It may be more ideological (maybe economic rationality?) than good sense. While many governments suggest that CT ensures transparency, the practice of CT does not ensure such a claim is necessarily valid, as details of tender review and assessment are rarely published and claims of cost savings cannot often be verified. There are suggestions that the transactions costs of CT compared to negotiation are as much as 10% higher.

The comments above have a pedigree in the literature on institutional maturity, and what we have through the current green initiative, as highlighting the issues, is effectively a relatively immature market where we have much to learn about how best to transition into and deliver post-transition to clean energy services, a cost efficient and cost-effective bus service. This is a good argument for negotiation rather than tender. ‘Optimal’ regulatory schemes when there is investment uncertainty aligns well with the theoretical arguments presented in Laffont and Tirole (1993).  Competitive tendering is a high-powered regulatory scheme, and under uncertainty[1], if risks cannot be reduced, it will increase the cost of capital. A lower powered incentive scheme, such as a negotiated contract (or rate of return regulation in the regulatory literature) may be optimal in this scenario, at least for a transition phase until uncertainty is lower. This would transfer risks to the government or users (depending on who assumes the financial consequences of unexpected cost changes) and implicitly assumes that governments (or users) are better able to absorb these risks, as they should, given that it is their policy commitment being implemented. As the product of joint decision making, negotiated contracts are peculiar insofar as both parties voluntarily surrender some measure of freedom to each other, and yet this sacrifice enables both to gain more than what would otherwise be possible until the risk are unambiguous and appropriately allocated in the determination of an efficient contract price.

As a caveat, we are not assuming that regulators are highly professional and skilled in the relevant tasks and responsibilities, any more than bus operators are, or that they both strictly comply with public accountability and transparency and low corruption, although these latter concerns appear to be more relevant to developing economies. The negotiated transition is a way of sharing skills that are greater in one group than another including other participants such as asset suppliers (both manufacturers and energy generators and distributors who have far more knowledge on green technology than other parties (at least to date)). Any presumption that negotiated contracts may not be optimal and deals are often done, even when there is uncertainty regarding the operational environment compared to competitive tendering, is not justified from the evidence (see Hensher 2015).

Regardless of the relative merits of these two procurement models, the re-tendering model is premised on the (often implicit) assumption that we have a high level of knowledge of the expected levels of costs associated with the provision of bus services in the next contracted period, and with stable technology this was easier to predict. Both the regulator and operators (incumbent or otherwise) have always had access to information that allows the establishment of a cost-efficient cost as a total cost/kilometre. With a stable technology, this can be calculated by assuming that the bus fleet will continue to be diesel with a known depreciation profile (typically straight line) over an agreed life of the bus asset[2]. In addition, the maintenance program centred on diesel vehicles is well established as are the requirements for staffing to ensure that the timetabled and other bus services are delivered to the market without delay. This also has been achieved with a clear appreciation of the size of the fleet required to fulfil contracted services. Under competitive tendering, the regulator has been able to receive bids and to compare them in such a way that they are strictly comparable in terms of what will be offered, essentially a well-designed level playing field (with known technology and associated costs). That is, the specification of all deliverables is very homogeneous, unambiguous, and deemed to align with best practice. The difference between winning and losing is effectively linked to cost comparisons and offers of improved service quality. Hensher et al (2016) investigated disruption costs in bus contract transitions and provided evidence that evaluators on tender evaluation committees do recognise the inherent risks in changing the service provider in bus contracts, and that it is possible to quantify the financial trade-off that evaluators make in balancing the risk associated with transition and disruption and the offer price. For example, if we take the median marginal rate of substitution between changeover cost and offer price reported by Hensher et al (2016), the prices offered by a new provider might be adjusted upwards by the evaluation committee in their recognition of the impact of uncertainty due to expected risk of incurring transition costs from a change of incumbent, with the adjusted amount depending on the lowest offer price[3]. This is exactly the risk setting that will exist (and worse) under the transition to a green fleet in particular.

Regardless of whether a CT model or a NPBC model has merit in the eyes of the regulator and/or the operator, the defining revelation that has surfaced over many years is in the detail of exactly how a bus contract should be structured and delivered, not only at the time of a bid or negotiation, but during the entire period of an active contract leading up to a new bid or negotiation. We are now in a very strong position to look back and review the elements that have worked and have not worked, not always as agreed by one or more parties in the contract chain, but rather what the more experienced regulators and operators have garnered from active participation. In many ways, the selection of a
CT or NPBC model is not the main concern; rather, the important consideration is the detail contained in a contract, and how it is interpreted by each party both under normal operations and when there is a dispute.

Some of the key elements that need drastic reform include a move to more flexible (hence less rigid) contracts, identifying risk and ensuring it is shared across all who gain to benefit, simplifying contracts at the ex ante bid stage with a recognition of an ability to review and revise during the tenure of the successful bidder, the opportunity to migrate to a collaborative contract (of great value in the decarbonisation transition), and  to protect trust in the partnership between a principle and an agent.

Whatever reform is undertaken, we should always be reminded that the the primary strategic objective is the long-term delivery of high-quality value for money public transport. In the reset journey, always think STO: Strategic (S), Tactical (T), Operational (O), a framework to capture the intent of the bus contract reset roadmap. The STO framework serves the process extremely well regardless of the modal or institutional setting.  It is a powerful way of establishing order in the planning and policy process. At the Strategic level the focus is on the establishment of broad goals and objectives and guidance on ways of achieving outcomes consistent with such goals (‘what do you want to achieve’). The Tactical level highlights the supporting mechanisms to achieve the strategic goals (where the contract and regulator sit); and The Operational level focuses on delivering the desired services to the market consistent with the strategic intent and aided by tactical mechanisms. Importantly, user and wider community are key stakeholders, with roles to play in clarifying goals at the S level, suggesting system design factors at the T level, and operational ideas at the O level, including feeding back on outcomes.

The strategic focus for promoting and improving public transport is very much about reducing external costs, supporting agglomeration economies and providing a mobility safety net (a merit good argument) and all tactical and operational initiative must have regard to this overriding set of objectives (see Stanley and Van de Velde 2008), something that is too often neglected in the details associated with contract design and delivery. Pushing governments to be more specific about the value it expects from public transport, would be a great step forward, since it adds substance on what value for money looks like. This then defines expectations for the scope of public transport service contracts, including what incentive mechanisms might need to include, and more clearly puts contract details in their right place (at T/O).   We will never get the T/O right if we do not spell out the strategic intent more fully. At present, we suggest that bus contracts are relatively good at focusing on the here and now, except when the market changes, and not at the future - there is too much conservatism without real long-term benefits assessed against the strategic objectives; and generally, contracts will have a schedule of services to be delivered on day one of the contract and sometimes, but rarely, will they have some future service changes containing aspirational changes for network enhancement in the future.

Footnotes

[1] Another reason why tendering may not be appropriate when there is much uncertainty and bidders value of the good are correlated is that the result may be prone to the "Winner's curse". We suggest this is part the result of poor tenderer assessment practices that fail to contrast the ultimate winners bid price with a benchmarked sense of ‘sensible prices.

[2] Operators in many countries have reduced the period of write off in line with their revenue flows in order to reduce the tax impost in a particular year; however, with a flat revenue stream there is little gain in so doing. In cases where operators continue to own buses, general requirements are for an average age of 12 years and no bus to be greater than 25 years. In NSW, for example, the government specifies an approved panel from which operators may purchase their buses. Individual chassis and body manufacturers tender to be placed on that panel.

[3] For example, if we have a bid offer from a non-incumbent of say $AUD6/km, then we might adjust this up to $6.25/km. as the relevant comparison with offer from an incumbent.

References

Hensher, D.A. (2015) Cost efficiency under negotiated performance-based contracts and benchmarking for urban bus contracts –are there any gains through competitive tendering in the absence of an incumbent public monopolist?, (presented at the 13^th International Conference on Competition and Ownership of Land Passenger Transport (Thredbo 13), Oxford September 15-19 2013) Journal of Transport Economics and Policy, 49, Part 1, January, 133–148.

Hensher, D.A., Ho., C. and Mulley, C.M. (2016) Disruption costs in contract transitions, Research in Transportation Economics,  59, 75-85. http://www.sciencedirect.com/science/journal/07398859/59

Hensher, D.A. and Stanley, J.K. (2010) Contracting regimes for bus services: what have we learnt after 20 years? Research in Transportation Economics, 29, 140-144.

Laffont, J.J. and Tirole, J. (1993) A Theory of Incentives in Procurement and Regulation. The MIT Press, Cambridge.