«Valentina Licata. Office of Rail Regulation, 1 Kemble Street, London, WC2B 4AN 4th September 2013 Dear Ms. Licata, GB Railfreight Ltd. response to ...»
Network Rail contracted Arup to re-calibrate the Capacity Charge for CP5 and the final report was published on 24th May 2013. As a result of this calibration Network Rail has concluded that the Capacity Charge rates for freight should increase by some 400% (from approximately £4 million per annum to £21 million per annum). This large increase, if applied using the current methodology, i.e. to all trains would have resulted in an unaffordable increased cost to FOCs. This therefore exacerbated the need to reconsider the existing structure of the Capacity Charge for CP5, particularly in light of its inconsistency with the Regulations.
RFOA proposal Original proposal based on Schedule 8 benchmark adjustments The RFOA made a proposal to Network Rail in August 2012 to incorporate the Capacity Charge within Schedule 8 benchmarks. This mechanism is already in use for the freight operator element of the benchmark, which is adjusted annually when there have been increases in total train miles on the network.
At that time Network Rail said there was insufficient time to consider this proposal for CP5 implementation.
RFOA is still of the view that adjusting the benchmarks for Network Rail as well as FOCs would be the most appropriate way of ensuring that Network Rail were correctly incentivised to accommodate additional traffic and suggests that further consideration is given to this in preparation for CP6.
Subsequent proposal based on marginal additional trains only
As an alternative the RFOA made an alternative proposal to the ORR in April 2013.
The basic proposal is that a baseline of freight train miles is established based on the same year as is being used to calibrate Schedule 8 benchmarks. At the end of each year this is compared to the actual freight train miles operated in that year by all freight train operators. Assuming this is a positive number the difference is multiplied by the capacity charge rate per train mile and then is charged to each freight train operators in proportion to the total number of freight train miles that they have each operated (ensuring no discrimination between operators). If the number of train miles has not increased over the baseline or has reduced the payment would be zero. The mechanism is in practice similar to other “wash-up” mechanisms such as the calculation of the EC4T charge, which is adjusted for actual use at the end of each financial year.
The same method of calculating the cost of the marginal impact is used based on the Arup report but the actual charge for freight is calculated on a marginal basis based on the number of train miles actually operated in the form of a wash-up at the end of the year.
This methodology supports the principle of paying for every new train mile operated on the network. It applies equally to all trains and is a transparent and simple with low administration costs.
Please see below example of calculation based on actual figures provided by Network Rail for 2012/13.
Zero based line This proposal would result in a baseline of zero payment (based 2010-12 train miles so in practice given increased mileage in 2012/13 there would be some payment) before growth (compared to approximately £4
million a year in CP4). The RFOA thinks this is not an unreasonable starting point for several reasons:
Payments made below the Schedule 8 benchmark level equate to an over-recovery of costs and are therefore a contribution to fixed costs, and therefore should be assessed against an ability to pay in line with the Access and Management Regulations.
In the CP5 draft determination the ORR has proposed changes to both the NR and the FOC benchmarks levels. According to the ORR’s calculations, at current performance levels the combined impact of these benchmark changes would be a reduction in Network Rail’s out payments on the freight Schedule 8 by £10.3 million. Therefore net of the reduction in Capacity Charge Network Rail’s net out payments would still reduce by £6.3 million assuming no change in actual performance levels.
32,000 32.0 30,000 30.0 28,000 28.0 26,000 26.0 24,000 24.0 22,000 22.0 20,000 20.0 :
The RFOA has however proposed a compromise that it is based on the same years as Schedule 8 benchmarks are set which is we now understand 2010/11 and 2011/12 which would result in a annual benchmark of 24,693,489 miles. In summary freight operators have contributed to reducing congestion on the rail network rather than causing additional congestion. Network Rail’s exposure to Schedule 8 payments has reduced, but a Capacity Charge has still been paid for all trains (freight operators have not been rewarded for their contribution to reducing congestion).
It could therefore be concluded that the train miles baseline was set at the 2001/2 level, when the Capacity Charge was last updated, i.e. a much higher level. This means that the efficiency gains already made by freight operators are incorporated into the new baseline.
Network Rail Payment Rate -Train Load Following the review of the Network Rail Payment Rate (the "payment rate") conducted during 1.
PROS, the payment rate was set at £17.47 per train minute of delay. The rate of £17.47 applied for the year 2009/10 and has since been uplifted annually for inflation. In the draft determination for PR13, the ORR propose to follow the same approach of annual uplifts for inflation such that the payment rate was £19.13 in 2012/13, is £19.74 in 2013/14 and would be uplifted for inflation in each year of CPS.
2. However, inflation is not the only factor that affects the per train minute cost of delay. Train load
- i.e. the amount/volume of goods moved -is also an important factor. As train loads increase, each train minute of delay affects more goods and inflicts greater costs on both freight operators and freight users.
3. The table below shows the elements of freight operator costs (sourced from ORR research), their relative sizes and how they respond to changes in train load: 1
4. The table above shows that for an increase in train load, 80% of the freight operator costs of delay would also increase proportionally.
5. The table below shows the elements of freight user costs (sourced from the AECOM/ITS report) and how they respond to changes in train load: 2
The table above shows that for an increase in train load, all freight user costs of delay would also 6.
7. In excluding changes in train loads from its calculations, the ORR is failing to compensate FOCs for increases in the consequences of delay. Since the entire premise of the payment rate is that it should compensate FOCs for the costs of delay and train loads are an important factor affecting those costs of delay, the payment rate should be adjusted to account for changes in train load.
We therefore suggest that the proposed payment rate should be adjusted for changes in train load 8.
since the beginning of CP4 and that, going forward, the payment rate should be adjusted annually to account for both inflation and changes in train load. In particular, the tables above demonstrate that freight user costs should change proportionally with average train load and that freight operator costs should change at 80% of the rate of the average train load.
9. Network Rail does not publish figures for the amount/volume of goods transported on the railway network; however, it does publish figures for the weight of goods transported. Although it is the amount/volume of goods that ditecdy affects costs of delay, the weight of goods acts as a reasonable proxy for the amount/volume of goods. One proviso to this is that the different commodity types have different densities and so using industry-level figures for changes in average train weight will not accurately represent changes in the amount/volume of goods moved.
10. Network Rail figures show that average train loads, as measured by tonnes of cargo (i.e. net of the weight of the rolling stock itself) per train, have increased at an average rate of 3.4% per annum between 2009/10 (the beginning ofCP4) and 2011/12. 3 Given the slight commodity shift towards intermodal during CP4, we believe that the average rate of 3.4% in fact masks a stronger increase in the amount/volume of goods moved per train. Consequendy, the true increase in annual volume of goods per train would be higher than 3.4% p.a. However, since there has only been a slight shift in commodity mix during CP4, we use the figure of 3.4% as a proxy for the increase in amount of goods transported but note that it is lower than the true rate for the increase in amount of goods transported for these years.
11. Official figures for average tonnes per train are not available for the years after 2011/12, but the trend of increasing average tonnes per train is forecast by Network Rail to continue throughout CPS. Since Network Rail's forecast for freight traffic in total tonne kilometres is not based upon average weight per train, dividing Network Rail forecast tonne kilometres by forecast train kilometres would be misleading due to significant forecast changes in commodity mix.
12. Both track access charges and increasing network congestion incentivise freight operating companies to increase train loads rather than the number of train movements. Furthermore, the Network Rail forecast appears to assume unconstrained demand growth; this would suggest Network Rail under-estimates the growth in average train load as freight operating companies face very real constraints on their ability to add extra train movements. For these reasons, we have used the historical growth rate of 3.4% in our following indicative analysis. 4 Network Rail Long Term Planning Process (April 2013) NR forecasts set out in Network Rail Long Term Planning Process- Freight Market Study Draft for Consultation, April2013
13. The table below sets out out proposed methodology for recalculating the payment rate from the beginning of CP4 and throughout CPS. First, we separate the payment rate of £17.47 at the beginning of CP4 into a freight operator cost component and a freight user cost component (taking the freight operator cost figures from ORR Researchl Secondly, we uplift the freight operator cost component for (i) inflation and (ii) 80% of the change in average train load.
Thirdly, we uplift the freight user cost component for (i) inflation and (ii) the. change in average train load. We then repeat each step on an annual basis.
ORR Review of Access Policy 20 I 0, Annex C
14. Using this methodology to correct the Network Rail payment rate for changes in train load gives an indicative payment rate in 2013/14 of £22.49 rather than £19.74 as currendy in place. By the end of CPS, further increases in train load produce an indicative payment rate of £30.57 as opposed to £22.77 and an average increase in payment rate during CPS of £5.61. The difference in payment rates reflects the significant extra costs of delay incurred due to increases in train loads which should be factored into the payment rate.
15. We note that the table above uses industry-wide (i.e. not corrected for differences in density of commodities) figures for average train weight growth for the years 2009/10 to 2012/13 and an estimate of industry-wide average train weight growth of 3.4% to calculate the payment rate for the years after and including 2013/14. When using actual figures rather than forecast figures to set future payment rates, the ORR should beware that, due to forecast changes in commodity mix, growth in tonnes per train is likely to under-estimate growth in the true driver of user costs which is the amount of goods being carried per train.
16. In conclusion, the ORR proposes in the Draft Determination that the current payment rate, as set at the beginning of CP4 and subsequendy uplifted for inflation, continue to be uplifted for ONS (RPI reference CHAW); Oxford Economics (ONS, Haver Analytics)
inflation during CPS. However, smce the beginning of CP4 train loads have increased at an average rate of 3.4% and are projected to continue increasing throughout CPS. As shown m the tables in paragraphs 3-S, train load is an important factor affecting the costs of delay per train minute because almost all cost consequences of delay are linked to the amount/volume of goods that are delayed. If the payment rate is to compensate freight operators for the costs of delay, it should therefore be uplifted to account for the increase in train load.