Moving towards the preferred option

This section draws on Sections 3 to 7 – the consideration of the 5 cases and makes an overall recommendation for the delivery of the scheme, specifically the appropriate reimbursement rate and then considers the overall value for money of the intervention.

Choice of scheme design, reimbursement rate and impact on costs

The Socio- Economic case laid out the estimated demand impacts of the scheme based on a range of assumptions, including the price sensitivity of bus demand, the level of fares and the potential for a larger scale impact on long-distance journeys. The demand impact is considered independent of the reimbursement rates set as it depends primarily on the actions of passengers (or potential passengers) who are not impacted directly by the level of reimbursement to operators.

Where the reimbursement rate may impact instead is supply – if the reimbursement rate is not generous enough to encourage participation or in the case of increased demand operator are unwilling to amend services then supply (of £2 fare capped services) may be constrained or there may be capacity issues that mean that higher level of increased demand may not be realised.

The costs of the scheme are significantly and directly dependent on both the demand response and the reimbursement rate. In terms of demand, if there is no demand uplift for an operator if the reimbursement rate is below 100% then operators will be worse off (See BRIA section of socio-economic case up). This means that the first recommendation of this OBC, in order to achieve agreement to a reimbursement rate less than 100% particularly for smaller operators serving remoter communities is:

In the core scenario the range of costs of the scheme, as they vary with reimbursement rate are as follows:

Table 35 – Core costs by reimbursement rate

Core Elasticity Scenario	Lower	Low	High
Increase in demand	9.5%	19.1%	38.2%
Core Cost	£85m	£85m	£85m
90% reimbursement	£84 m	£91 m	£106 m
95% reimbursement	£89 m	£96 m	£112 m
100% reimbursement	£93 m
£102 m	£118 m
Neutral reimbursement rate	91%	84%	72.5%

Note that the core cost is the estimated amount that the scheme will cost before there is a demand reaction and is equivalent to leaving operators no better or worse off.

All things being equal, a standard approach would use the neutral reimbursement rate associated with the mid-range of the demand response estimates. This would equate to the low scenario and a reimbursement rate of 84%.

However, all other things are not equal and there are an number of further factors to consider.

Asymmetric impacts on operators

Not all operators will face equal demand impact. In fact it is almost certain that each operator will experience the scheme differently. As such a mid-point even if 100% accurate will leave some operators worse off and it is not unreasonable to suppose that these will be smaller operators concerned with less profitable or subsidised services that provide key local connections.

Although the clause in recommendation 1 will mitigate this effect, it is possible that larger losses in the short term, before the mechanics of the clause can come into effect, could place significant financial pressure on operators and in the worst case for operators currently at the margin cause them to fail.

At the same time, it is a clear fact that 100% reimbursement would leave any operator who experienced an increase in demand better off.

This asymmetry suggests that there is value in choosing a rate between the “optimum” rate but below 100%.

Existing pilot scheme in Highlands and Islands

The existing pilot scheme has a reimbursement rate of 90% on the mainland. Given that there are services between the two areas, offering a different rate whilst the pilot is running would be operationally complex specifically and particularly for those services that are of significant passenger volume and cross both boundaries ie Glasgow to Inverness and Glasgow to Oban.

There is also some potential risk that operators who are operating only in the HITRANS (mainland) area may challenge a higher rate elsewhere in the country.

Uncertainty over demand response estimates when faced with large changes in price

Standard demand response analysis tends to deal with relatively small changes in price i.e. annual fare increases of around 10% excluding inflation. This suggests that the higher estimates, even if appropriate for small price changes, may overstate the demand response when faced with large price falls as it requires significantly greater numbers of individuals to change their behaviour and those individuals may have a range of non-price factors (comfort, convenience etc.) that may overrule price based decision making. This is in line with standard economic theory – with many forms of demand curve, the price elasticity changes along the demand curve and movements down the demand curve reduce the sensitivity of demand to price.

In addition there is a view that Covid-19 may have had a permanent downward impact on the price sensitivity of demand which is not yet reflected in current estimates.

Capacity constraints

The High scenario is associated with demand changes of almost 40%. This is likely to cause severe capacity (supply) issues. As such there is an argument that such levels of demand increase will not be realised in practice. This casts some doubt as to using the High scenario to define a mid-point.

Taking these 4 points together there is clear evidence and rationale for offering a reimbursement rate in excess of the 84% suggest above as “theoretically optimum”. Looking at just the Low and Lower scenarios would suggest a rate of 87.5% but considering the asymmetric impacts and specifically the current rate in Highlands, would strongly suggest that a 90% rate is reasonable.

Table 36 – Core cost by recommendation

Core Elasticity Scenario	Lower	Low	High
Increase in demand	9.5%	19.1%	38.2%
Core cost (before demand change)	£85 m	£85 m	£85 m
90% reimbursement	£84 m	£91 m	£106 m

This results in estimated core scenario costs of between £84m and £106m with a central estimate of around £90m. The following section explores the VfM of these recommendations.

VfM assessment of option

The estimated annual costs of the recommended option for bus are outlined in the tables below.

Table 37 – Estimated annual bus costs and rail costs

Core Elasticity Scenario	Lower	Low	High	Worst Case
Increase in demand	9.5%	19.1%	38.2%	38.2%
Core cost (before demand change)	£85 m	£85 m	£85 m	£110m
90% reimbursement	£84 m	£91 m	£106 m	£150 m

The additional estimated costs for rail

Rail Scenario	Low	Medium	High	Worst Case
Rail impact (-ve)	£25 m	£35 m	£45 m	£60 m

There are estimated administration costs of £200,000. These are treated separately as part of the VfM analysis that follows. This means that the costs of the scheme in total are estimated (annual rounded up to nearest £5m) to be, focusing on the low and high (the 2 central) scenarios:

Table 38 – Overall costs

Overall Cost	Low	High
Bus cost	£90 m	£105 m
Rail	£25 m	£45 m
TOTAL	£115 m	£150 m

It is recommended that the costs are considered as a range due to the significant uncertainty. If a point estimate is required then the most sensible approach is to take the mid-point of the entire range of £90m to £150m i.e. £120 m.

The VfM analysis of the scheme needs to be made in the context of the mechanism by which it functions – operators being compensated at 90% in order that passengers pay lower fares. This is technically a transfer payment – an exchange of money rather than an economic cost – money flows from Transport Scotland to operators to consumers and balances out overall.

Perhaps counterintuitively, the VfM does not depend on the amount paid to operators – any payment made to operators is neutral – it is either a benefit to consumer or, in the case where demand increases above the point where operators are left no worse off, a benefit to both consumers and operators.

In addition, it is appropriate to apply “distributional analysis” to weight the benefits to bus users upwards (given the analysis above) based on their income characteristics. In summary:

If demand does not increase

The scheme has a negative impact equal to the administration costs of £200,000 before distributional impacts are considered. However, any reasonable distributional weighting will result in positive benefits and positive VfM overall.

If demand increases

Then without distributional weighting, a very small increase in demand will more than offset the administration costs before distributional weighting is applied and there will be positive VfM overall. Applying distributional weighting will likely increase these benefits further.

In addition, any demand increases involving shifting from private car will have associated carbon (and wider congestion) benefits, further reinforcing the VfM of the scheme. This will be a careful focus of the evaluation.

Whilst this VfM analysis suggests that the scheme will show in formal terms that it is value for money, it should be remembered that the prime purpose of the scheme is to reduce the costs of public transport for users.

Long term impacts

There are a number of potential long-term issues with the extension to SPT, and ultimately to the rest of the country, that will need to be addressed before there is a full national scheme.

Future removal of bus market

The key issue is that the current reimbursement model relies on compensating operators based on their Adult Single fare. Once the scheme is introduced there will no longer be an unconstrained (by the price cap) market and no future commercial Adult Single fares.

Adjustments to current levels can be made in the short term to reflect changing costs after the scheme has been running for a year, for example. However, in the long-term it is likely that an alternative reimbursement scheme will be required. This scheme will allow the collection of data to inform that development.

Impact on Concessionary Fare reimbursement model

A side-effect of the scheme is an additional impact on the current reimbursement model for existing Concessionary Fares. This is also based on a proportion of existing commercial Adult (and Child) single fares. As with the scheme itself, an ultimate national rollout will mean the removal of this market and will apply similar issues to that discussed above to wider Concessionary travel.

Impact of inflation

Over, time the impact of inflation is likely to have a significant impact on the costs of the scheme. This is due to the mathematics of the cap. By way of example:

Suppose the average fare in an area is £2.50.

Currently reimbursement would be 90% of £0.50 or 45p per journey. If the fare on which reimbursement is payable to operators were to increase by 10% to £2.75, reimbursement would increase to 68p per journey ie a cost increase of 50%.

This means that maintaining a fare cap at £2 will become significantly more expensive over time.