Scotland-wide Older and Disabled Persons Concessionary Bus Scheme - Further Reimbursement Research
7 Recalculation of the Reimbursement Rate
7.1 The National Reimbursement Rate
7.1.1 The principal stimulus for the current research was the need to have a systematic method for updating the National Reimbursement Rate so that it continued to satisfy the Scottish Government's "no better off, no worse off" objective.
7.1.2 Fare levels, prices and costs all change over time, and it has been particularly evident that the average shadow fare, based on the adult cash single fare that concessionary passengers would be charged for their journeys, has been increasing at a much faster rate than prices generally, and indeed the bus fares experienced by non-concessionary passengers. Such increases in fares are bound to increase the gap between the concessionary journeys actually made, and those that would be made if those fares had to be paid by passholders, i.e. will increase generation.
7.1.3 The calculation process illustrated in the preceding chapter demonstrates how the logical consequences of changes in fares can be followed through in terms of changes to the allowance for generation, and hence the calculation of reimbursement for both revenue forgone and additional costs.
7.1.4 We have identified a range of elasticity parameters that can be used to update the Reimbursement Factor. We have examined the basis for calculating an overall average discount factor, and also for updating the additional cost element of reimbursement. The illustrations of the reimbursement calculations provide a basis for setting a revised Reimbursement Rate for 2013-14, taking into account the strength of the evidence base, theoretical issues and also the practical consequences of change.
7.1.5 However, it should be noted that although this updating process will boost the likelihood that the National Reimbursement Rate delivers "no better off, no worse off" reimbursement at a national, "average operator" level, it will not address a fundamental weakness that a national rate will almost inevitably not leave bus operators no better off, and no worse off at the individual level:
- bus operator fares vary widely, and consequently the percentage of concessionary journeys generated by the concession will also vary;
- the commercial policies that operators pursue will lead to significantly different discounts being provided to passengers by some operators compared to others; and
- operational characteristics will vary, and so will the likely justification for payment of additional costs.
7.1.6 The consequence is that a national rate that does not take into account of these differences will over-reimburse some bus operators and under-reimburse others. This in itself is potentially a source of conflict with EEC regulations intended to regulate state aid.
7.1.7 A more practical problem is that since reimbursement is largely driven by the cash single fare, a national rate provides a strong commercial incentive for individual bus operators to increase their cash single fares at a greater rate than they would do otherwise. The fact that the cash single fare has risen as rapidly as it has (having increased by 27% in real terms between 2006-7 and 2012-13, as measured by the CPI, and 25% as measured by the RPI) is almost certainly one consequence of this.
7.1.8 For the individual operator, a national rate that is not influenced by the operator's own fares means that as an operator increases its cash single fare, concessionary revenue increases in direct proportion. In contrast, with non-concessionary passengers market forces operate to reduce revenue yield since demand is moderated by passenger resistance. This incentive to increase the cash single fare will have been accentuated by the imposition of caps on annual reimbursement totals, leading some operators to have intentionally brought forward fare increases to maximise their share of a limited annual pot.
7.1.9 In our view, a move away from a single national reimbursement rate would therefore reduce the rate of increase in fare levels, (and hence reduce the rate of growth of reimbursement payments) that has been experienced in the last few years, and reduce the need for reimbursement caps that impose an arbitrary ceiling on expenditure.
7.1.10 Alternatives to the National Rate are inevitably more complex than the present system. The great attraction of the current system is its simplicity, which in operational terms means that individual operators can have a clear idea about future revenue streams, while (particularly with smartcard-based automation) minimising Government's administrative overhead. If a system that was more sensitive to individual bus operator characteristics was to be put into place, care would therefore be needed to ensure that an appropriate balance was struck between "theoretical" correctness and practical issues.
7.1.11 Operator specific calculation of the Reimbursement Factor need not be particularly complex or onerous. For example, it would be easy to calculate operator-specific reimbursement factors from the data that is already collected. This would reduce incentives to excessive fare increases, without incurring significant additional demands for data or administrative resources. But addressing other differences between operators (e.g. with regard to additional costs) could impose substantial additional complexity that might be disproportionate to benefits.
7.1.12 Even though the focus of much of the current research has been on updating the National Reimbursement Rate, the methods developed for calculating reimbursement could largely be applied at a more disaggregate level without significant further research.
7.1.13 However, it is important to note that a radical change to the reimbursement system could have a number of unintended consequences that were destabilising to the bus network and create dis-benefits. The economics of bus operations vary greatly, and a one-size-fits-all reimbursement solution might lead to some significantly sub-optimal outcomes. Because care is required, a change to operator specific rates is not an option for the short term, but this should not inhibit consideration of changes to the current National Scheme arrangements in the longer term. We recommend that Transport Scotland undertakes some preliminary investigation of potential options and their consequences.
7.2 Practical steps with regard to future reimbursement rates
7.2.1 On the basis that 2013-14 reimbursement arrangements in Scotland are on the same principle as currently, with one reimbursement rate applied to all operators, Transport Scotland needs to decide on the specific reimbursement rate itself so that reimbursement payments for individual operators can be calculated from 1 April 2013.
7.2.2 Although we have stated our preference for the specific reimbursement parameters quoted above in Chapter 6, we recognise that following consultation with CPT and other interested parties, Transport Scotland may choose a different set of parameters. We strongly recommend that the Transport Scotland decision is articulated in terms of explicit choices of reimbursement parameters, which should be regarded as fixed for 2013-14, and explicit assumptions about price levels and fare levels in 2013-14.
7.2.3 The reimbursement parameters that need to be specified are as follows:
- the Discount Factor, as defined in Chapter 3. For the avoidance of doubt, this should be prior to the application of any assumptions about generation, to clearly separate the concepts of average fare estimation from the estimation of generation;
- the elasticity parameters, i.e. values for Lambda and Beta as defined in Chapter 4, and with the Beta value dimensioned to a given price level measured by a given price index;
- definition of the price index to be used to convert the average fare forgone at the "current" price to the price level at which Beta is dimensioned; and
- the base additional cost rate, specified at a given point in time and price level.
7.2.4 The National Reimbursement Rate can then be calculated as an output from a calculation involving these reimbursement parameters, and the explicit assumptions about 2013-14, namely:
- the average shadow fare in the year, in nominal (current price) terms;
- forecasts of the level of the chosen price index; and
- forecasts of the price index to be used to inflate the additional cost rate.
7.2.5 It is recommended that assumptions are chosen following consultation with CPT and other interested parties. The starting point needs to be clearly established base-line numbers at specified dates in the base year e.g. the out-turn average shadow fare at an agreed date, and likewise for the price indices.
7.2.6 Were serious consideration to be given to moving away from the shadow fare as the basis for monitoring operator fare levels, appropriate adjustments might be needed to the discount factor, but in principle this should not affect the overall quantum of reimbursement that is calculated.
7.2.7 Given these inputs, calculation of the Reimbursement Rate itself can be carried out in one step with one formula. However, it is easier to understand the logic involved if the arithmetic is separated into distinct steps. Appendix C provides a worked example of a practical reimbursement calculation that may be helpful in presenting proposals for change.
7.2.8 Calculation of an appropriate Reimbursement Rate for 2014-15 would follow an identical process to that described above. Of the various reimbursement parameters used in 2013-14, we recommend that the discount factor is subject to regular review, requiring some process for gathering in equivalent data on commercial bus ticket sales and revenues as provided to the current study by CPT. We recommend that there is a formalisation of the requirement on operators to provide this data. The calculations set out in Chapter 3 provide a template that can be used to inform such a review. Consideration should be given to using the "Method (b)" approach to help review assumptions about the journeys made using period tickets that are implicit in discount factor calculations.
7.2.9 With regard to the elasticity parameters, we do not believe that the elasticity of passenger responses will change significantly over time and so a detailed review of these need only be carried out periodically, for example if and when any significant additional evidence on concessionary elasticities becomes available. We recommend that Transport Scotland monitors emerging research on concessionary travel, and any relevant developments in England and the other devolved administrations.