7 Incremental Cost and Revenue Scenarios
7 Incremental Cost and Revenue Scenarios
7.1 Overview of Approach
7.1.1 The approach taken here, in overview is as follows:
Incremental Costs
- calculate vessel acquisition costs;
- calculate fuel and other costs based on service frequency in each scenario and vessel specification;
- calculate pier dues associated with volumes from projected demand (see below);
- calculate berthing dues based on service frequency and vessel specification; and
- estimate the total incremental costs for all Growth and Vessel scenarios.
Incremental Volumes and Revenue
- estimate total route volumes based on the four whole route demand projections (see Chapter 6);
- estimate uplift in total route volumes as a result of additional services on the route (elasticity based);
- estimate market share for the proposed new foot-passenger and passenger and vehicle services - based on the number of vessels operating the route (and hence the timetable achieved);
- estimate the resulting incremental revenue from outturn average revenues per person / car / cv; and
- estimate the total incremental revenue for all Growth and Vessel scenarios.
7.1.2 Incremental costs and revenues are then compared to assess the feasibility of each growth / vessel scenario.
7.2.1 If a new town centre vehicle carrying ferry service is introduced, vehicle-based Cowal residents and visitors would then have a choice between the two routes. There are only a relatively limited number of parameters which could determine this choice:
- crossing time;
− crossing times between Cowal and Western Ferries were the same at 20 minutes, despite the town centre route being longer; - length of the operating day;
- crossing frequency;
− will depend on the number of vessels in service - average fares paid;
- access journeys times to terminals, relative to the journey being made;
− clearly differs by broad origin / destination - convenience of journey (ticketing arrangements etc); and
- vessel comfort and facilities.
7.2.2 A spreadsheet-based choice model has been developed to represent how users behave in the face of these choices, based on a Discrete Choice modelling framework. The model represents the journeys between census datazones in Dunoon and beyond and key locations to the south of the Clyde for Dunoon residents, and vice versa for visitors to Cowal. It includes a representation of access times by datazone, fares, crossing times and service patterns across the day.
7.2.3 This choice modelling clearly requires the current Western Ferries service to be represented as well as the new town centre service. No change to current Western Ferries timetabled services has been assumed as part of this exercise, although this issue is explored further in Chapter 8.
7.2.4 As has been noted, the key element of this exercise is the impact of moving from a foot-passenger only service to passenger and vehicle service. For example a two-vessel foot-passenger service must be compared with a two-vessel passenger and vehicle service to establish an incremental difference (assuming all other things remain the same, ie the same timetable is being run). By definition, the number of foot-passengers would be the same between these two scenarios (as the service is identical from the perspective of a foot-passenger) and further consideration of foot-passenger impacts can be discounted here beyond the initial foot-passenger only subsidy calculation.
7.2.5 Proximity to each ferry terminal is clearly a key issue for Dunoon residents in choosing which ferry service to use. Accessibility analysis undertaken as part of this study has shown that Hunter's Quay is the closest ferry terminal for 29% of local residents and Dunoon town centre is the closest terminal for 71% of local residents.
Fares
7.2.6 Fares is also a key issue here. Cowal Ferries data from 2010 has been used to estimate the outturn revenues per passenger, car and CV, as was shown in Chapter 5. These figures are very much lower than the published fares due to the level of local discounting available (typically up to 50%). This means that published fares clearly cannot be used as a basis of estimating potential future revenue streams.
7.2.7 However when these implied fares were applied to Western Ferries volumes, the revenue figure derived was higher than Western's reported revenues. This suggests that the average fares paid on Western Ferries in 2010 were lower than for Cowal Ferries, despite the apparent similarity in the published 'standard' fares. The implied fares were 17% lower than Cowal average fares. This is most probably a reflection of the mix of users on each route at that time.
7.2.8 Cowal Ferries would be the first ferry service most visitors would come across when travelling from the east into Gourock. Indeed, the 2007 survey data suggested that a greater proportion of Cowal Ferries users (33%) were 'visitors' (as opposed to Cowal residents), the equivalent figure for Western Ferries being 26%. Visitors are clearly much more likely pay a full fare on the ferry, perhaps being unaware of the opportunities to buy cheaper tickets at local retail outlets, and clearly being far less likely to buy discounted multi-journey tickets. Local Dunoon residents are clearly much more likely to purchase discounted fares.
7.2.9 The choice of fares is key here as they determine: (i) the mode share likely to be achieved on the town centre service; and (ii) the incremental revenue increase in moving from foot-passenger to passenger and vehicle. The question which this poses for the choice modelling is which set of fares to use? The choices are:
- 2010 outturn Cowal Fares: £2.27, £5.76, £31.48 (passenger / car / cv); or
- 2010 estimated Western Fares: £1.88, £4.78, £26.13 (passenger / car / cv).
7.2.10 Given that (i) any new town centre service would have to attract market share from Western Ferries (and thus be price competitive); and (ii) the profile of these users will reflect the current Western Ferries user profile, it would seem more appropriate to assume Western Ferries outturn fares for the purposes of this exercise. However both sets of fares are included initially for added clarity.
7.2.11 The (estimated Western) fares used in calculating revenues have been adjusted to 2013 values to be consistent with Pier and Berthing Dues using RPI (January 2013), an increase of 12.8% on January 2010 (giving values of £2.12, £5.40, £29.33 for passengers, cars and commercial vehicles respectively).23
7.2.12 Note that under the current formulations, any application of Road Equivalent Tariff (RET) on the town centre route would result in higher fares than those noted above. We have therefore not considered RET any further as part of this exercise.
Vessel Scenarios
7.2.13 Three different vessel scenarios are considered here, based on the acquisition of two, three or four new vessels.
7.2.14 A two-vessel operation on the route would be able to provide a 30 minute service with two sailings per hour in each direction, providing the vessels had sufficient maneuverability etc to achieve this. Single vessel operation later in the evening would provide opportunities for vessel maintenance.
7.2.15 Three vessels would allow a 20 minute service with three sailings per hour across much of the day with the service dropping to two vessels at quieter periods.
7.2.16 Although Western Ferries operate four vessels, a 20 minute frequency is generally operated across the day, with half hourly services in the early morning and evening. Only at certain times of the day or to accommodate flexible peaks in demand is the fourth vessel deployed. At present, only Saturday mornings and Friday afternoons see a timetabled 15 minute service. At no point are four vessels per hour timetabled on Monday to Thursday. Four vessels would therefore be required to provide absolute parity with Western Ferries, and this scenario is also considered. Note that a four vessel scenario where all four vessels operate a 15 minute services has not been considered at this stage, as this would seem excessive.
7.2.17 These three different vessel scenarios (involving two, three or four new town centre vessels) give rise to three different timetable scenarios as noted above and these in turn would lead to differing market shares as described below.
Market Shares
7.2.18 A key issue with respect to market share is that, for those travelling between Dunoon and Glasgow or the east (the majority movement by far across the Firth), the town centre route provides shorter drive times on the south of the Clyde and somewhat shorter drive times for many on the north of the Clyde. As such, if all things were equal with respect to both ferry services, it would be reasonable to assume that a majority of car-based users would use the town centre service.
7.2.19 The 2007 surveys referred to in Chapter 5 suggested that 54% of car-based journeys made (across the two combined services) were between Cowal / Helensburgh / Lomond and the east (ie Renfrewshire, Glasgow and areas further east). A further 32% were between Cowal / Helensburgh / Lomond and Inverclyde itself.
7.2.20 The scale of these figures was confirmed by a set of indicative traffic count surveys which were undertaken on 18/12/12 between 1000 and 1200 at McInroy's Point and between 1400 and 1530 at Hunter's Quay.
7.2.21 At McInroy's Point, a total of 187 vehicles were counted disembarking from the ferry. Of these:
- 90% turned left ie towards the east - Gourock / Glasgow; and
- 10% turned right ie towards the west A78 / Ayrshire.
7.2.22 At Hunter's Quay, a total of 109 vehicles were counted disembarking from the ferry. Of these:
- 74% turned left ie towards Dunoon town centre; and
- 26% turned right ie onto the A815 northbound to Sandbank and beyond.
7.2.23 Overall, these figures suggest that the majority of vehicular traffic (at least in December) is travelling between Dunoon and the east towards Gourock and beyond.
7.2.24 Note that Hunter's Quay is around two miles from Dunoon Ferry Terminal and the distance between McInroy's Point and Gourock Ferry Terminal is around 2.5 miles. The route east from McInroy's Point is currently through Gourock town centre which is subject to traffic congestion at peak times.
7.2.25 As an illustration, a two-vessel service running in addition to Western Ferries would account for 40% of the total crossings on the route (assuming no change in current Western Ferries timetable). As such, if fares were identical it could reasonably be expected to achieve around a 40% market share. However, the large proportion of eastbound car-based travel noted above (towards Glasgow) would mean that a greater market share would be expected (as the 2.5 mile trip through Gourock town centre is avoided if using the town centre route). Similarly a three vessel service would account for 47% of all sailings and a Western Ferries parity scenario would clearly provide 50% of all sailings.
Forecast Market Shares
7.2.26 The choice model has been calibrated to reproduce the 10% car market share achieved by Cowal Ferries in 2010, based on the outturn fares and timetables of that time. A 'constant' has been included as part of this calibration process to represent the restricted nature of the previous Cowal operation, including the inconvenience of the previous ticketing arrangements which was a prominent theme in the consultation and reportedly a deterrent to the use of that service. It is assumed that this constant is removed in future scenarios, ie a more flexible ticketing arrangement would be in place, which is not tied in to ticketing practices across the CalMac network. The restricted operating day is also assumed to be lifted.
7.2.27 The 2010 Cowal vehicle ferry timetable accounted for around 25% of total sailings between Gourock and Dunoon. As such, a market share of around 25% could have reasonably been expected based on this frequency, if all other things were equal. Removing the constant modelled in the 2010 base scenario gives a projected market share of 20% on the town centre route, ie had the various restrictions not been in place, a 20% market share could have been achieved in 2010 with the single vessel, hourly operation. Indeed there is a precedent as CalMac did achieve this level of market share around 2000. This level of market share gradually reduced over the years with the greater overall convenience offered by Western Ferries.
7.2.28 Table 7.1 below shows the projected market share achieved with the removal of this constant and the introduction of a two-vessel, three-vessel and Western Ferries timetable parity scenario (assuming current Western Ferries timetabled services), based on the choice modelling process.
7.2.29 As noted above, the balance here is mainly between service frequencies and the drive times on either side of the Clyde, ie for most, the town centre route would offer shorter drive times. The choice modelling exercise suggests that, with identical fares, even with only two vessels, the inconvenience of a less frequent service is outweighed by the shorter access times for many journeys and a market share of around 56% is predicted. This rises to 64% if a Western Ferries timetable were to be adopted.
7.2.30 Like any forecast, these figures should be treated with some caution. The forecasts were derived from the best data available at the time, but forecasts of this nature can always be improved with better data etc. In the analysis which follows, in addition to using these projected market shares, we have also estimated the market share required to realise incremental revenues greater than incremental costs, ie the 'tipping point' for feasibility.
7.2.31 The Table also clearly illustrates how sensitive the choice between operators would be to fares. The forecast market shares are far lower if 2010 Cowal outturn fares are used. If the two competing ferry services were to become very similar in other respects (ie frequency, crossing times etc), then fares would become the key determining factor for most users. The high market shares estimated above are based on fares parity with Western Ferries. These (lower) fares will of course provide lower revenues per trip made and this is explored further below.
7.2.32 As such, if the two services were to converge in terms of their offering, the prospective market share would be very sensitive to small changes on the supply side, most notably fares.
7.3.1 This section brings together the cost data and the projected market shares into a 15-year financial appraisal. No discounting has been applied and all values are expressed in today's prices.
Revenue
7.3.2 Any new transport service which is introduced typically takes time to build up patronage to its full forecast level ('ramp-up'). This was seen on the Gourock-Dunoon route itself when Western Ferries commenced operation in the 1970s. Table B12.1 in the rail-based Passenger Demand Forecasting Handbook (PDFH) provides recommended values for these time lag effects in the context of new rail services. In the absence of an equivalent data set for the ferries sector, PDFH values have been applied here as follows:
- Mid year 1: 60% of forecast;
- Mid-year 2: 78% of forecast;
- Mid year 3: 90% of forecast; and
- Year 4 onwards: 100% of forecast.
7.3.3 The combination of estimated market shares and average fares paid per passenger / car / CV produces revenue estimates for the prospective new service. As noted previously, the fares used to estimate the revenues are based on Western Ferries 2010 estimated fares factored up by RPI to 2013, and these are: £2.12, £5.40 and £29.35 for passengers, cars and commercial vehicles respectively. These fares therefore include the impact of Western Ferries not charging SPT Card holding passengers a passenger fare (whether foot-passengers or vehicle based passengers). The analysis which follows implies that any new operator would adopt the same fares structure and therefore not charge vehicle based SPT Card holding passengers. No real terms change in fares has been assumed for forecast years.
7.3.4 As an example, for Scenario 1 (Static Demand) in 2015, a forecast 56% share of the total vehicle based market would eventually generate around £4.3m per annum in incremental revenue (799k vehicle based passengers * £2.12 + 363k cars * £5.40 + £22k commercial vehicles * £29.35). Figure 7.1 shows the forecast annual incremental revenue for vehicle based travel on the town centre service, based on a two-vessel service. Note that in 2009 and 2010, total route revenue was around £7.5m. The 'lagged' build up in revenue is evident here.
Figure 7.1 Two vessel annual incremental revenue estimates, by demand scenario
7.3.5 The total 15-year incremental revenue for each Scenario and vessel combinations is shown in Table 7.2 below. The annual average figure is also shown for reference.
7.3.6 These figures confirm that revenues rise moving from Scenario 1 to 2 and 3 then fall with the declining scenario (4). Within each scenario, revenues rise as the service moves broadly from two sailings per hour to three then to more than three in the case of the Western Ferries parity scenario.
7.3.7 The large majority of revenue would come from a transfer from Western Ferries.
Incremental Costs
7.3.8 The key issue here is the incremental cost change in moving from a foot-passenger service to a passenger and vehicle service operated by the same number of vessels running the same timetable.
7.3.9 The analysis presented in Chapter 4 noted that the significant cost increases are associated with:
- Vessel acquisition costs (Foot-passenger vessel circa £3,000,000, and passenger and vehicle vessel £6,000,000): for the purposes of this analysis it has been assumed that these vessels are financed @ 4.5% per annum, and the value is written off at the end of the 15-year period24;
- Fuel: the passenger and vehicle ferry will have a higher fuel consumption by virtue of being a larger vessel;
- Ship maintenance and repairs: again a vehicle carrying ferry would be more expensive to maintain and repair;
- Spares and stores;
- Dry docking costs for refit;
- Berthing dues (see below);
- Pier dues - all additional vehicles and passengers carried will be liable for Pier Dues when embarking and disembarking at both harbours (see below); and
- Insurance.
7.3.10 Note that the analysis has suggested that no additional crew would be required to operate the passenger and vehicle service, relative to a foot-passenger only service, although the case is borderline as discussed in Chapter 4.
7.3.11 The incremental cost estimates for fuel, ship maintenance and repairs, spares and stores, dry docking and insurance are shown in Table 7.3 below.
Berthing Dues
7.3.12 Berthing Dues are payable on Gross Tonnage berthing at each harbour and the frequency of these calls, ie the fees are independent of the number of passengers or cars carried. Current Berthing Dues are as follows:
- Gourock: £0.33 per GT and
- Dunoon: £0.078 per GT.
7.3.13 Discounts are applied on berthing dues at Gourock as laid out in Table 7.425.
7.3.14 No discounts are currently published for Dunoon.
7.3.15 The number of calls at each port is clearly dependent on the number of vessels being operated to provide the service and the resulting running timetable. The number of calls per annum at each harbour has been estimated as:
- Two vessels: 10,731 calls;
- Three vessels: 14,600 calls; and
- Western Ferries parity: 16,164 calls.
7.3.16 Median GTs of 355 and 800 for the foot-passenger and passenger and vehicle vessels respectively have been assumed (see Chapter 4). The following berthing dues would apply in 2017, and all years thereafter (based on today's rates). Table 7.5 shows the berthing dues for the two vessel service.
7.3.17 The incremental cost of providing a like-for-like passenger and vehicle service is therefore around £649k per annum, based on the 2017 pricing schedule. This value is slightly lower at £590k and £619k in 2015 and 2016 respectively.
7.3.18 Table 7.6 shows the same data for a three-vessel scenario.
7.3.19 The incremental increase in Berthing Dues is around £811k per annum under this scenario in 2017. The equivalent values for 2015 and 2016 are £753k and £782k respectively.
7.3.20 Finally Table 7.7 shows the incremental berthing dues associated with a service providing parity with the current Western Ferries published timetable.
7.3.21 In this case, the incremental increase in berthing dues is £877k per annum. The equivalent figures for 2015 and 2016 are £819k and £848k. These figures are based on Western Ferries published timetable - the actual number of sailings operated by Western is higher in practice due to additional ad hoc sailings at times of high demand, so higher berthing dues would be incurred to match this total.
7.3.22 Note that the incremental costs therefore do not rise in direct proportion to the number of sailings due to the discounting at Gourock.
7.3.23 The total 15-year incremental berthing dues figures are therefore as follows:
- Two-vessel scenario: £9.6m;
- Three-vessel: £12.1m scenario; and
- Western Ferries scenario: £13.1m.
7.3.24 Note that these figures depend crucially on the Gross Tonnages of the vessel, and this aspect of vessel design is subject to considerable uncertainty. A lower differential in gross tonnage would reduce these figures but a higher differential would increase incremental berthing dues. This is explored further in Chapter 8.
Pier Dues
7.3.25 The current published Pier Dues tariffs are shown in Table 7.8 below.
7.3.26 Until 31 March 2013, the Berthing Dues discounts noted above also applied to Pier Dues at Gourock. These discounts therefore ceased as of 01 April 2013 and full payment will be due. These Pier Dues clearly represent a significant proportion of revenues received from paying passengers.
7.3.27 An estimate of the 15-year incremental Pier Dues payable based on the above market shares and volumes is shown in Table 7.9 below, based on the current published rates.
7.3.28 Pier Dues therefore account for around 45% of fares revenue. Of this Gourock accounts for around 63% of total fees with Dunoon accounting for the other 37%.
Summary of Incremental Costs and Revenues
7.3.29 A summary of the estimated incremental costs and revenues is shown in Table 7.10 below. The table shows the forecast incremental costs and revenues associated with each vessel scenario combined with each overall route growth scenario. The net incremental revenue is then shown for each vessel / scenario combination. Positive numbers (ie incremental revenues outweigh incremental costs) are highlighted in green and negative numbers (incremental revenues are less than incremental costs) are highlighted in pink. The projected market share is also shown for each vessel scenario for convenience (based on fares parity with Western Ferries).
7.3.30 Also shown is the 'tipping point' in each case. When the net revenue is positive, this tipping point indicates the market share where revenue ceases to be positive. When the net revenue is negative, this tipping point indicates the market share where revenue becomes positive. For example in the Scenario 1 / Two-vessel case, a forecast 56% market share results in a net impact of +£7.8m over the 15-years (ie the incremental revenue outweighs the incremental costs). However if the market share was to fall below 42%, the incremental costs would outweigh the incremental revenues, ie the service configuration would have to achieve 42% to break even.
Table 7.10 Overall 15-year incremental costs and revenue summary
7.3.31 It can be seen that a large majority of the incremental costs associated with moving from a foot-passenger service to a passenger and vehicle service are attributed to increased Pier Dues (higher traffic levels) and Berthing Dues (vessel with a higher GT). These two elements account for around 70% of the total incremental cost in a two-vessel scenario.
7.3.32 Table 7.10 therefore shows that a two-vessel scenario is a feasible option in all four of the forecast growth scenarios, based on the assumptions made here and the definition of feasibility used, ie the net revenue impact is positive. Note that this net revenue would be liable to corporation tax26. Under the two vessel scenario, this equates to a return of £7.7m (Scenario 1 -Static), £9.0m (Scenario 2 - Gradual Recovery), £13.6m (Scenario 3- Trend growth), or £3.4m (Decline), over the 15-year period.
7.3.33 A three-vessel scenario is also feasible for all growth scenarios except Decline (Scenario 4). However the Western Ferries timetable parity scenario would only be feasible under Scenario 3 (Trend Growth), the scenario with the highest growth of the forecast scenarios considered here.
7.3.34 It can reasonably be assumed that Scenarios 2 and 3 are the most likely outcomes in terms of total route volumes, assuming economic recovery becomes established. Also, under these growth scenarios, there is a reasonable 'cushion' in terms of the forecast market share and the market share required to 'break even' on the two-vessel and three-vessel options.
7.3.35 This analysis has been undertaken using the best available data and assumptions based on best professional judgement where required. Inevitably, the outcomes will be sensitive to these underlying assumptions and a total of 12 scenarios have been considered here. The 'tipping point' whereby the service ceases to be, or becomes viable has also been reported to provide as full a picture as possible. .
7.3.36 The key uncertainty in this analysis is the potential for competitive response from Western Ferries. Given this uncertainty we have assumed no competitive response here, but as the level of service on the two competing ferries converges, small changes to fares, running times etc would be expected to have a significant impact on market share, and therefore the feasibility of running a passenger and vehicle service with no increase in subsidy.
7.3.37 In the above, it is assumed that the current infrastructure could be used in the short term for any new ferry service, although future infrastructure investment would be required in the medium term.
7.4.1 The focus of this report is on the impact of moving from a foot-passenger service to a passenger and vehicle ferry service, and whether this requires additional subsidy or not. This has been explored fully above.
7.4.2 However, the absolute subsidy implied by the foot-passenger service is also clearly important. As has been noted, the foot-passenger vessel specification undertaken here has been focussed on matching the weather related reliability performance of the previous Streaker vessels, and thus the vessels specified would be anticipated to have higher costs than the existing passenger vessels on the route.
7.4.3 Table 7.11 below provides an initial estimate of the subsidy requirement based on the absolute costs and revenues associated with the foot-passenger service as specified here, for each vessel and growth scenario. An allowance of 5% has been included for an operator margin, although clearly this figure would be subject to the outcome of any tendering process.
7.4.4 Note that a single passenger vessel scenario has not been considered as there are currently two vessels on the route.
Table 7.11 Foot-passenger Service Subsidy Estimates
7.4.5 It can therefore be seen that a two-vessel foot-passenger service would require an annual subsidy of around £2.9m to £3.1m. The current Argyll Ferries contract, at the time of award amounted to around £1.7m per annum - based on much smaller vessels. In contrast, the previous Cowal Ferries service received an annual subsidy of more than £3m (however it was operating vessels certified for 500 passengers and higher crewing levels).
7.4.6 Any vessel scenario involving more than two vessels would require a substantially greater subsidy.