Decision No. LET-R-66-2010 - Review of the Railway Interswitching Regulations

LET-R-66-2010

April 21, 2010

File No. 7360-6

Dear Madam/Sir:

Re:  Review of the Railway Interswitching Regulations

1. Introduction

Further to its consultation initiated in December 2007, the Canadian Transportation Agency (Agency) is soliciting further comments as part of its ongoing review of the Railway Interswitching Regulations (Regulations).

The Agency indicated that a specific objective is to review the provisions of the Regulations to determine whether they accurately and effectively reflect current railway operating practices and the costs of providing interswitching services within the statutory framework prescribed by section 128 of the Canada Transportation Act (CTA).

In particular, in an effort to ensure that the prescribed interswitching rates are established at an appropriate level, the Agency proposed to implement interswitching rate changes that were reflective of the Agency's estimated interswitching costs for the year 2007. A total of 12 submissions were received from the stakeholders in response to the consultation. For your reference, the submissions received are available for consultation on the Agency's Web site at http://www.otc-cta.gc.ca under the heading "Consultations".

A number of issues were raised by stakeholders. After consideration of the arguments made by stakeholders advocating changes, the Agency determined that there would be no changes made to the Regulations with respect to: deregulation; redefining car blocks; expansion of the interswitching limit; applicability of interswitching rates to Class II and III railways; and performance and efficiency concerns.  The Agency's position on these issues is set out in Appendix A.

However, after considering the other comments received and the significant changes observed in railway operating practices since the original proposal in 2007, the Agency determined that  its publication/methodology for the development of the interswitching service costs requires review to ensure that the resulting rates accurately reflect current interswitching costs incurred by the railway companies.

Based on this review, it was determined that a number of changes to the publication/methodology used by the Agency to determine costs as well as specific cost factors should be changed. These are:

• an increase in the level of contribution towards railway constant costs;
• changes in the publication/methodology for the determination of the interswitching variable costs; and
• a reaffirmation of how movements made under co-production agreements and irregular operating practices are treated.

Applying these changes to the Agency's publication/methodology and to cost factors, as well as updating cost estimates, from the current 2002 base to 2009, the interswitching rates for trains of 1 to 59 cars would increase interswitching rates for all zones except Zone 4. The rates would decrease for all zones for block trains of 60 cars or more.

Given the substantive nature of the proposed changes, the Agency has decided to carry out a second consultation and seek comments before finalizing interswitching rates and proceeding to amend the current Regulations.  Further details on these changes and the proposed interswitching rates are set out below.

2.1 Contribution toward railway constant costs

Since the enactment of the CTA on July 1, 1996, which promulgated a new rate prescription directing that rates established by the Agency be "commercially fair and reasonable" to all parties, the Agency has considered that a contribution of 7.5 percent over variable costs represented an appropriate compensation for railway constant costs (also referred to as fixed costs) in respect of interswitching movements. Accordingly, the rates were amended in 1997 to include a new contribution level of 7.5 percent.

The issue of the magnitude of the contribution toward railway constant costs has generated diametrically opposed views among stakeholders in past consultations. The users of interswitching services have generally supported the continued use of a contribution level of 7.5 percent over railway variable costs on the grounds that there was no apparent reason to vary the earlier Agency determination.

Conversely, the providers of interswitching services have generally maintained that they should receive a compensation that provides for full recovery of their constant costs. Railway companies have argued that as a fundamental principle, when regulatory intervention is deemed necessary, such intervention should be made in such a way as to least distort the marketplace. They also argue that a level of compensation for such an imposed public duty that is less than the total cost incurred by the railway company to perform the service would be a distortion of the marketplace.

The Agency, through its mandate as an economic regulator, performs various economic determinations which incorporate a contribution towards railway constant costs. These applications use a system-wide average contribution to constant costs which is derived from empirical data produced by the railway companies and verified by Agency staff.

In its review of this issue, the Agency is of the opinion that it must be consistent in applying contribution towards railway constant costs for all its relevant cost and rate determinations.  In making determinations on the compensation of a regulated rail activity, the Agency must be fair and reasonable to all parties.  The Agency does not see any convincing rationale for continuing to apply a different, lower contribution towards constant costs for interswitching rate determination. Therefore, to standardize the contribution towards railway constant costs, the Agency considers it appropriate to treat the compensation of the interswitching services as it would other regulated activities. This consistency in the Agency's approach will provide the rail carriers with rates that provide "commercially fair and reasonable" compensation for costs incurred and will still support effective competitive access through interswitching.

Accordingly, the Agency proposes that the contribution level included in the interswitching rate structure be adjusted to reflect the average system contribution to constant cost at a level, calculated by Agency staff, at 20.3 percent of the variable costs.

2.2 Methodology for the determination of the interswitching variable costs

In general, interswitching costs used to calculate the rates are determined as follows.

The average variable costs used for the determination of the proposed interswitching rates are based on the actual costs of the interswitching movements for each zone, at all interchanges where interswitching is performed.

For each such interchange, the Agency determines the costs associated with the specific work activities required for each of the distance zones. The cost estimates are derived by measuring the time required to perform both general work activities that are common to the handling of the interswitching of traffic and specific activities that relate to the pickup or delivery of the interswitching traffic. The resulting average costs by yard are then weighted by the traffic interswitched at each yard, based on the distribution of the interswitching traffic, to produce a system average cost per car for each zone.

Finally, the respective system average costs for both the Canadian National Railway Company (CN) and the Canadian Pacific Railway Company (CP) are adjusted for inflation and estimated changes in railway company productivity and then averaged, based on the interswitching traffic of each railway company in that particular zone, to generate a system average cost measure per car for each of the four distance zones.

The measures of time required to perform the work activities are updated by site inspections of selected yards. Different yards are visited each year so that, over time, most of the yards have been inspected. These inspections focused primarily at capturing recent changes in the operations and the efficiency at handling the traffic volume to be interswitched at a particular yard.

As a result of these visits, switching times in specific yards may be adjusted to reflect observed changes in operations and traffic volumes. Generally, changes in operations result in more efficient operations and lower interswitching times.

The interswitching costing publication/methodology does not necessarily produce uniform variations for each distance zone because it is significantly interdependent on the distribution of the interswitching traffic patterns.

For instance, shifts resulting in less traffic at a low cost yard or more traffic at a high cost yard can negatively impact on the cost estimates and contribute to higher interswitching average costs. Similarly, shifts in the distribution of the traffic patterns among the distance zones may also affect the interswitching cost levels.

Specifically, decreases in traffic volumes in a specific zone can result in increased average interswitching times, as the time that the yard crew spends switching cars is spread over fewer cars, thus increasing the average time per car and ultimately the interswitching costs for that specific zone.

Finally, changes in the railway companies' unit or operating costs impact on the determination of the interswitching costs. Therefore, there are several factors that come into play in the determination of the railway interswitching costs that can contribute to the variation in the estimated costs.

A review of the empirical data, which point to some significant operational changes since 2003, suggested the need for three changes to the publication/methodology used to calculate interswitching rates: discontinuing the use of linear regression; using single-year results in place of multi-year averages and using a more accurate cost publication/methodology for interswitching block trains.

2.2.1 The use of linear regression

Linear regression was previously used in the development of interswitching rates to smooth out the results so that the rates increased proportionally with an increase in distance from the interchange.  The midpoint distance for each zone was regressed against the zone level costs per car, and weighted based on volume of carloads.  That regression found the "line of best fit" through the weighted data points.

Beginning with the rates in this proposal, Agency staff have eliminated the use of linear regression as it forces a relationship which the current data does not support.  Rates do not necessarily increase proportionally with increases in distance from the interchange.  The reasons for this include the fact that the rail network is composed of different grades of track and that customers and their sidings are not identical.

For example, the proposed rate for Zone 4 has decreased. This is significant because approximately one-third of the carload volume interswitched in trains with fewer than 60 cars takes place in Zone 4. The decline in the proposed rate occurred because traffic originating or terminating in Zone 4 is concentrated at a limited number of interchanges and the geography and the operational conditions prevailing at these interchanges, and their associated rail yards, are such that their work activities and costs are either similar or lower than the system average cost for Zone 3 traffic.

Accordingly, the current data does not validate the existence of a linear relationship between the interswitching cost of a movement and the distance travelled. As a result, it was decided not to use a regression technique to impose a linear relationship where empirical evidence shows one does not exist, in respect of trains of 1 to 59 cars.

2.2.2  Single-year results versus multi-year average

In previous cost determinations, the Agency has based its assessment of the interswitching variable costs, in respect of trains of 1 to 59 cars, on a three-year moving average of the traffic counts in order to minimize the effect of the variations in the traffic distribution patterns and ultimately reduce the variability of the results. In the present proposal, the process was modified to use only data relating to the actual traffic interswitched from the most recent year available. This change would allow the analysis to more closely capture the evolving operational environment and respond to observed material changes in the work activities and ultimately reflect more accurately the costs of interswitching.

Therefore, the Agency has replaced its previous process of averaging the work activities observed during the most recent two visits for any given yard by using only the most recently observed information.

2.2.3 Cost publication/methodology for interswitching block trains

The Agency introduced a new category of traffic for block trains of 60 cars or more in 1988.  Since then, it has used a simulation model for the estimation of the variable costs, which generated the theoretical work activities required to perform the interswitching of block trains. However, over the years, the interswitching of block trains has increased significantly both in terms of traffic volume and the number of users and locations involved, so that the volume of traffic currently interswitched exceeds 100,000 cars annually and is performed in all zones. Given the larger population of block interswitching traffic, the Agency can now rely on actual observed work activities to develop the variable costs associated with this category of traffic. This change increases the accuracy of the cost assessment of block train interswitching.

2.3 Co-production agreements and non-conventional handling arrangements

In this interswitching cost determination, the Agency proposes to continue to include the variable costs of interswitching traffic that has been moving under the terms of co-production agreements and other non-conventional handling arrangements^{Note 1}. These data are included where appropriate for both trains of 1 to 59 cars and for block trains of 60 cars or more.

The implementation of the co-production agreement between CN and CP governing the traffic originating or terminating in the Vancouver area resulted in a significant variation in the physical handling of the traffic, notably for interswitching services.   However, the Agency considers it appropriate to continue to include interswitched traffic moving to or from Vancouver under the terms of the co-production agreement in its calculation of variable costs.

The traffic moving under the terms of the co-production agreement is not transferred at the historic interchanges in Vancouver, but is transferred at a location that the railway companies deem more operationally efficient. For the purpose of the cost determination, the traffic is considered to be transferred at CN's Thornton and CP's Coquitlam yard facilities in Surrey and Coquitlam, respectively. The transportation costs associated with the segment east of the yards are covered under the co-production agreement, and are, therefore, not considered part of the interswitching movement. Although the railway company determines the best way to move the traffic, the Agency is of the opinion that shippers continue to be entitled to regulated interswitching if they are within 30 kms of the interchange.

3.  Interswitching rate proposal

Consistent with the current interswitching rate structure, the interswitching costs were developed for trains of 1 to 59 cars and for block trains of 60 cars or more in each of the four interswitching zones.  In addition, the costs per each additional kilometre beyond Zone 4 were also computed for these two categories of interswitching movements.

A contribution towards fixed costs of 20.3 percent of variable costs was then added to the variable costs to establish the proposed interswitching rates. The resulting rates were rounded down to the nearest dollar to ensure that the effective contribution does not exceed the proposed level of 20.3 percent.

Table 1 illustrates the proposed rate changes for each interswitching zone for interswitching less than 60 cars as well as for interswitching a block of 60 or more cars. The table presents, for comparative purposes, the current regulated rates, which became effective November 5, 2004 and are based on the interswitching variable cost estimates for the year 2002 and a contribution towards fixed costs of 7.5 percent, and the present proposal which has been developed on the basis of the 2009 interswitching variable cost estimates and the increase in the level of the contribution to railway constant costs to 20.3 percent.

The assessment of the variable costs associated with trains of fewer than 60 cars produced diverging results. The variable costs associated with Zones 1 to 3 increased from their level in 2002, contributing to an increase in the proposed rates ranging from 18.3 to 24 percent.

Conversely, the variable costs for Zone 4 declined, resulting in a reduction in the proposed rate of 20.3 percent. Coincidentally, this figure is the same as, but unrelated to, the figure for the level of contribution to constant cost.

The change in the proposed rate for Zone 4 as noted earlier in section 2.2.1, is the result of a significant portion of the total traffic being concentrated in a limited number of relatively efficient interchanges.

4. Impact of the rate proposal

Stakeholders are reminded that current rates have been in place for a period of more than 6 years without any change.

Based on the 2007 traffic distribution patterns, the rate proposal would result in a global increase of 4.2 percent for the movements of 1 to 59 cars, which should generate a railway company revenue increase of approximately $ 1.57 million for this category of traffic. However, as the proposal also includes a general reduction of 16.7 percent in the interswitching rate structure for blocks of 60 or more cars, it is anticipated that this change would produce a reduction in railway company revenues derived from this category of traffic in the magnitude of $ 1.63 million.

It is proposed that the new interswitching rates will be in effect from the date of approval of the amended Regulations for a period of five years. However, the Agency will continue to monitor railway operations and cost factors and will consider on an annual basis whether any adjustments to the rates are warranted.

5. Consultation

The Agency requests your views on this regulatory amendment proposal. In particular, the Agency is seeking your comments in respect of the specific impact that the proposed changes may have on your transportation services and options. The Agency intends to review all submissions received and will take into consideration the points of view expressed in making its final determinations for proposed amendments to the Regulations that will be published in  the Canada Gazette.

Your submission should be forwarded, no later than May 21, 2010, to:

Canadian Transportation Agency
Ottawa, Ontario
K1A 0N9
Attention: John Corey
Facsimile: 819-953-5564

or by electronic mail to: rif-rir@otc-cta.gc.ca

These Regulations are available on the Internet at http://laws.justice.gc.ca/en/SOR-88-41. Submissions received during the course of this review will be available for consultation on the Agency's Web site at http://www.otc-cta.gc.ca.

Additional information can be obtained by contacting John Corey at 819-953-9930, 888-222-2592 (toll free) or TTY 1-800-669-5575.

Sincerely,

Cathy Murphy
Secretary

BY THE AGENCY:

• ________________________________
• Member
• ________________________________
• Member

APPENDIX A

Issues not subject to this consultation

1. Deregulation 

   The Agency is a quasi-judicial tribunal tasked with the administration of the Canada Transportation Act (CTA).  Any change to the concept of regulated interswitching would require a change in policy which is under the purview of Transport Canada, and any change to the CTA would have to be initiated by Parliament, not the Agency.

2. Redefining car blocks

   During past consultations, several parties proposed that the Agency examine the feasibility of establishing four rate categories for multiple car block sizes of 1 to 24, 25 to 49, 50 to 99, and 100 or more cars. Few sidings can accommodate blocks of 60 rail cars or more.  Sidings with a capacity for 25 or 50 rail cars are more common.  Establishing smaller car blocks would allow smaller shippers to take advantage of lower car block rates.

   However, railway companies have been generally opposed to any change to the car block sizes from the current 60-car threshold.

   The substantial rate reduction for interswitching movements of 60 cars or more, is brought about because, unlike  movements of smaller consists, the interswitching of blocks of 60 cars or more consists of a "hook and haul" type of movement that does not require any classification or marshalling activities in the yard. Smaller proposed consists which would move through yards would not enjoy a similar cost reduction, nor consequently a lower block interswitching rate.

   The Agency examined the interswitching movements performed in 2007 in terms of the quantity of cars interswitched in each movement.  Almost half of the cars interswitched (48.2 percent), fall into two categories at either end of the spectrum, consists of 9 cars or less and consists of 120 cars or more. Consists varying from 30 to 59 cars represented only 10.3 percent of the total number of cars interswitched. The creation of new categories of blocks for consists of 30 to 59 cars would be of limited benefit to the shipping community, affecting only 10.3 percent of cars and 6.6 percent of shipper facilities.

   Also, extracting the most efficient movements from the current rate category for trains of 1 to 59 cars, would inevitably induce an increase of the costs for the new category composed of consists between 1 and 29 cars. Although this measure would be beneficial to a limited number of shippers, the derived benefit would be made at the detriment of a majority of shippers (90.4 percent) which would face an increase in the rate for consists of 1 to 29 cars.

   A similar /conclusion can be reached in respect of the creation of a new block category for consists of 100 cars, for example, which account for 36.8 percent of the total number of cars interswitched, but is generated by a limited number of shippers/facilities estimated at 8.

   Given the above considerations, the Agency does not propose, at this time, to amend the existing rate structure to allow more categories of car blocks.

3. Expanding the interswitching limit 

   The 30 km limit was implemented as urban areas grew, so that the practical focus on reducing rail line congestion in urban areas and providing competitive opportunities for shippers would continue. The current interswitching limit is appropriate at this time.

4. Applicability of interswitching rates to Class II and III railway companies

   The majority of interswitching movements are performed by Class I railways. The costs and publication/statistics used to calculate the regulated interswitching rates are those of the Class I railway companies.  Although the regulated interswitching rates may not be appropriate in the view of short line railway companies, a deviation from the current costing and rate setting exercise, in the view of the Agency is not warranted nor practical.

5. Service performance and efficiency concerns 

   Service performance and efficiency concerns are a level of service issue, and are not dealt with through the Regulations, but rather through the application of the level of service provisions of the CTA.