SURFACE TRANSPORTATION BOARD DECISION DOCUMENT
    Decision Information

Docket Number:  
EP_646_1

Case Title:  
SIMPLIFIED STANDARDS FOR RAIL RATE CASES

Decision Type:  
Decision

Deciding Body:  
Entire Board

    Decision Summary

Decision Notes:  
DECISION MODIFIED THE BOARD'S SIMPLIFIED RAIL RATE GUIDELINES, BY CREATING A SIMPLIFIED STAND-ALONE COST APPROACH FOR MEDIUM-SIZE RAIL RATE DISPUTES AND REVISING ITS THREE-BENCHMARK APPROACH FOR SMALLER RAIL RATE DISPUTES. THE BOARD’S DECISION ALSO PLACES LIMITS ON THE TOTAL RELIEF AVAILABLE OVER A 5-YEAR PERIOD UNDER THESE TWO SIMPLIFIED APPROACHES.

    Decision Attachments

391 KB


Approximate download time at 28.8 kb: 6 Minutes

Note:
If you do not have Acrobat Reader, or if you have problems reading our files with your current version of Acrobat Reader, the latest version of Acrobat Reader is available free at www.adobe.com.

    Full Text of Decision

*****

38326 SERVICE DATE – SEPTEMBER 5, 2007

EB

This decision will be printed in the bound volumes of

the STB printed reports at a later date.

 

SURFACE TRANSPORTATION BOARD

 

DECISION

 

STB Ex Parte No. 646 (Sub-No. 1)

 

SIMPLIFIED STANDARDS FOR RAIL RATE CASES

 

The Board modifies its simplified rail rate guidelines, creating a simplified stand-alone cost approach for medium-size rail rate disputes and revising its three-benchmark approach for smaller rail rate disputes. The Board also places limits on the total relief available over a 5-year period under these two simplified approaches.

 

Decided: September 4, 2007

 

TABLE OF CONTENTS

Existing RATE REASONABLENESS STANDARDS................................................................ 7

Regulatory Framework.......................................................................................................... 7

Constrained Market Pricing................................................................................................... 8

Simplified Guidelines Adopted in 1996................................................................................. 10

Revised simplified guidelines......................................................................................... 12

I. Simplified-SAC Methodology..................................................................................................... 13

1. Objective....................................................................................................................... 13

2. Methodology.................................................................................................................. 15

II. Three-Benchmark Methodology.................................................................................................. 16

1. Comparison Group......................................................................................................... 17

a. Comparability Factors........................................................................................ 17

b. Selection Process............................................................................................... 18

2. RSAM Range................................................................................................................ 19

3. Method To Calculate RSAM and R/VC>180.................................................................... 19

4. Rate Reasonableness Presumption & Other Relevant Factors.......................................... 21

III. Procedural Matters................................................................................................................... 22

1. Procedural Schedules..................................................................................................... 22

2. Discovery....................................................................................................................... 24

3. Jurisdictional Inquiry....................................................................................................... 26

IV. Eligibility Criteria....................................................................................................................... 26

1. Relief Limits................................................................................................................... 27

2. Carrier Objections.......................................................................................................... 29

3. Litigation Cost Estimates................................................................................................. 30

4. Risk Factor.................................................................................................................... 32

5. Aggregation of Claims.................................................................................................... 32

6. Carrier Manipulation of Dispute...................................................................................... 33

Regulatory Flexibility Act............................................................................................ 33

CONCLUSION............................................................................................................................. 34

Appendix A – Road Property investment.................................................................. 38

1. Land.............................................................................................................................. 39

2. Roadbed Preparation..................................................................................................... 39

3. Track............................................................................................................................. 41

4. Tunnels.......................................................................................................................... 42

5. Bridges and Culverts...................................................................................................... 42

6. Signals and Communication............................................................................................ 45

7. Buildings and Facilities.................................................................................................... 46

8. Public Improvements...................................................................................................... 47

9. Mobilization, Engineering, and Contingencies................................................................... 48

Appendix B – Total SARR Operating Expenses......................................................... 49

I. Background................................................................................................................................ 49

II. Operating Expenses in Simplified-SAC Case.............................................................................. 49

Appendix C – Public Comments & Board ResponSeS............................................. 52

I. Simplified-SAC Approach.......................................................................................................... 52

1. Congressional Intent....................................................................................................... 52

a. Three-Tiered Approach...................................................................................... 52

b. Use of a SAC-Based Method............................................................................ 52

c. Simplified and Expedited..................................................................................... 53

2. Testing........................................................................................................................... 54

3. Necessary Requirements for Simplification...................................................................... 55

a. Use of Existing Infrastructure............................................................................... 55

b. Use of Existing Route......................................................................................... 56

c. Use of Existing Traffic Group.............................................................................. 57

4. Use of URCS Data for Operating Expenses.................................................................... 57

a. No Movement-Specific Adjustments................................................................... 58

b. Limited Adjustments........................................................................................... 58

c. Equipment Valuation........................................................................................... 60

d. Indexing of URCS Data...................................................................................... 60

e. URCS Variability Factor..................................................................................... 61

5. Use of Rolling Averages for Road Property Investment................................................... 61

a. High-Density vs. Low-Density RPI Costs............................................................ 61

b. Eastern Costs vs. Western Costs........................................................................ 62

c. Verification of Data............................................................................................. 63

6. Internal Cross-Subsidy Analysis...................................................................................... 63

7. Rate Reduction (Maximum Markup Methodology).......................................................... 64

8. Computation of Rate Relief............................................................................................. 65

9. Post-Prescription Follow-ups......................................................................................... 66

10. Discovery..................................................................................................................... 66

a. Disproportionate Burden..................................................................................... 66

b. Other Issues....................................................................................................... 68

11. Filing Fee..................................................................................................................... 69

12. Procedural Schedule..................................................................................................... 70

13. Results Compared to Full-SAC.................................................................................... 71

II. Three-Benchmark Approach...................................................................................................... 72

1. Union Pacific’s Objections to Three-Benchmark Approach............................................. 72

2. Reliance on the Mean..................................................................................................... 74

3. Other Relevant Factors................................................................................................... 76

4. Access to Waybill Sample.............................................................................................. 78

5. Changes to the RSAM and R/VC>180 Benchmarks.......................................................... 80

6. Comparison Group......................................................................................................... 82

7. No Adjustments to URCS.............................................................................................. 84

8. Regulatory Lag – Cost and Revenue Adjustments........................................................... 84

9. Waybill Sample Revenue Adjustments............................................................................ 85

III. Eligibility................................................................................................................................... 85

1. Summary of Public Comments........................................................................................ 85

a. NPRM Proposal................................................................................................ 85

b. Public Comments on Original Proposal............................................................... 87

c. Refinement of the Proposal................................................................................. 89

d. Supplemental Comments on “Small Claims” Refinement...................................... 90

2. Litigation Cost Estimates................................................................................................. 92

a. Simplified-SAC.................................................................................................. 92

b. Three-Benchmark.............................................................................................. 93

IV. Jurisdictional Threshold............................................................................................................. 94

V. Other Comments....................................................................................................................... 98

1. Exempted Traffic............................................................................................................ 98

2. Cross-Border Traffic...................................................................................................... 99

3. Class II and Class III Carriers...................................................................................... 101

4. The Kansas City Southern Railway Company............................................................... 102

5. Mandatory Mediation................................................................................................... 103

 


BY THE BOARD:

 

In this rulemaking, the Board seeks to make its rail rate dispute resolution procedures more affordable and accessible to shippers of small and medium-size shipments, while simultaneously ensuring that the new guidelines do not result in arbitrary ratemaking. In 1995, Congress directed the Board to “establish a simplified and expedited method for determining the reasonableness of challenged rail rates in those cases in which a full stand-alone cost presentation is too costly, given the value of the case.” 49 U.S.C. 10701(d)(3).[1] In an effort to respond to this directive, the Board adopted the guidelines set forth in Rate Guidelines – Non-Coal Proceedings, 1 S.T.B. 1004 (1996) (Simplified Guidelines). A decade has passed, however, without any shipper presenting a case that has been decided under Simplified Guidelines.[2] The Board held public hearings in April 2003 and July 2004 to examine why those guidelines had not been used by shippers and to explore ways to improve them.[3] The Board heard the views of rail shippers, railroads, rail labor, state governments, and other parts of the federal government. In general, the shipper community perceives the existing guidelines as too vague, and as requiring prolonged litigation over whether a shipper even qualifies to use them.

The Board concluded that significant changes to Simplified Guidelines were necessary to achieve the dual statutory goals of providing captive shippers meaningful access to regulatory remedies for rail rates that are unreasonable, while recognizing the need for railroads to earn a reasonable return on their investments so that they will have the resources to make the investment needed to continue to serve the transportation needs of their customers. Therefore, in 2006, the Board proposed to (1) create a simplified stand-alone cost (Simplified-SAC) procedure to use in medium-size rate disputes for which a full stand-alone cost (Full-SAC) presentation is too costly, given the value of the case; (2) retain the “Three-Benchmark” method of Simplified Guidelines, with certain modifications and refinements, for small rate disputes for which even a Simplified-SAC presentation would be too costly, given the value of the case; and (3) establish eligibility presumptions to distinguish between large, medium-size, and small rail rate disputes.[4] Initial comments on this proposal were submitted in October 2006, reply comments in November 2006, and rebuttal comments in January 2007. A hearing was held on January 31, 2007, and final comments were received in February 2007.

These proposals received mixed reviews. In general, the railroads favor the Simplified-SAC approach, urging the Board to either abandon the Three-Benchmark approach entirely or greatly modify it in a way that would complicate the approach. In contrast, shippers favor the Three-Benchmark approach, arguing that the Board should either abandon attempts to use a simplified SAC approach or delay implementing the Simplified-SAC proposal until it is tested. Shippers also urged the Board to dramatically increase the eligibility limits, to ensure that all captive shippers have an effective forum to pursue rate relief. Carriers oppose any increase in the proposed eligibility limits.

We conclude that both the Simplified-SAC and Three-Benchmark approaches should be made available to shippers. While neither approach offers as much precision and degree of confidence as a Full-SAC analysis, these alternative dispute resolution procedures address the concern that many shippers believe they cannot challenge their rail rates because the costs of litigation would exceed the amount in dispute. Shippers’ litigation costs in recent Full-SAC cases have approached $5 million. And while the reforms adopted in Major Issues in Rail Rate Cases, STB Ex Parte No. 657 (Sub-No. 1) (STB served Oct. 30, 2006) (Major Issues) may reduce those litigation costs, the reduction would not be large enough to make a Full-SAC case accessible to all rail shippers. The Simplified-SAC approach provides a reasonable means of retaining the advantage of a Full-SAC presentation, in simplified form: the ability to detect abuses of market power whereby a railroad forces a captive shipper to pay more than is necessary for the carrier involved to earn adequate revenues and thereby forces the captive shipper to cross-subsidize parts of the defendant’s rail network it does not use or benefit from. But shippers have offered persuasive evidence that even a Simplified-SAC presentation would likely cost up to $1 million to litigate. Captive shippers with rail rate disputes that cannot justify a Simplified-SAC presentation must have some forum for rate relief. So, the Three-Benchmark approach is needed to fill the gap that would otherwise exist in rail rate protections.

We are persuaded, however, that the proposed eligibility approach must be modified to ensure that all captive shippers have a meaningful forum for seeking protection from unreasonable rates. Accordingly, we will permit a captive shipper to select whether to pursue relief under a Simplified-SAC or Three-Benchmark approach, but place limits of $5 million on the relief available over a 5-year period under the Simplified-SAC method, and $1 million on the relief available over the same period under the Three-Benchmark approach. This approach follows the small claims court model used in civil litigation, a long-accepted alternative dispute resolution process whereby procedures and discovery are expedited, but with limits placed on the relief available. In this fashion, we will not try to assign a dispute to a particular simplified approach, which would require the Board to assess the value of the shipper’s case and analyze the potential merits of a dispute in advance of any evidence as to the reasonableness of the challenged rate. Rather, the shipper will evaluate the value of its own case and select the methodology that it believes is best suited for the amount at stake.[5]

Parties offered a wide variety of suggested refinements to the proposed Simplified-SAC and Three-Benchmark methods. We have reviewed those suggestions and implement the following refinements to the original proposal:

         Require parties to participate in a 20-day, non-binding mediation process at the outset of the case;

         Expedite the procedural schedules to the maximum extent practical;

         Address any request for revocation of an existing exemption before considering a related rate complaint under the simplified guidelines;

         For the Simplified-SAC analysis:

o       require the SAC analysis to focus on the primary route during the test year for the traffic at issue;

o       exclude depreciation on equipment when calculating operating expenses;

o       change treatment of any trackage rights expenses to better reflect the particular route analyzed;

o       revise how prior Full-SAC cases will be used to simplify the road-property investment analysis;

o       remove the annual adjustment process for a rate prescription;

o       increase the filing fee to $10,600.

         For the Three-Benchmark analysis:

o       provide access to the unmasked, confidential Waybill Sample upon the filing of a complaint, subject to the Board’s customary protective orders, but only for the traffic of the defendant carrier;

o       exclude traffic of a non-defendant carrier from the comparison group;

o       maintain the focus of the RSAM and R/VC>180 benchmarks on potentially captive traffic, but revise the way these two benchmarks are calculated;

o       presume that the challenged rate is unreasonable if it falls outside a reasonable confidence interval around the mean of the adjusted comparison group;

o       permit either the shipper or the railroad to submit evidence of “other relevant factors” to rebut the presumption;

o       exclude from the Three-Benchmark approach challenges to local rates of a shortline or regional carrier, but permit challenges under the Simplified-SAC approach.

In the body of this decision, we describe the two simplified approaches adopted here and the limits on relief available under each. With the exception of the modifications to our approach to eligibility, which are discussed in the body of this decision, we summarize in the appendix the public comments, our responses to them, and our reasons for the refinements to the original proposal.

Existing RATE REASONABLENESS STANDARDS

Regulatory Framework

Where a railroad has market dominance, its transportation rates for common carrier service must be reasonable. 49 U.S.C. 10701(d)(1), 10702. Market dominance is defined as an absence of effective competition from other rail carriers or modes of transportation for the transportation to which a rate applies. 49 U.S.C. 10707(a). The Board is precluded, however, from finding market dominance if the revenues produced by a challenged rate are less than 180% of the carrier’s “variable costs” of providing the service. 49 U.S.C. 10707(d)(1)(A). Variable costs are the portion of railroad costs that have been determined to vary with the level of traffic, using the Board’s Uniform Rail Costing System (URCS).[6]

Only the Board may determine if a common carrier rate is unreasonable. 49 U.S.C. 10501(b). When a complaint is filed, the Board may investigate the reasonableness of the challenged rate, 49 U.S.C. 10704(b), 11701(a), or dismiss the complaint if the complaint does not state reasonable grounds for investigation and action. 49 U.S.C. 11701(b). If, after a full hearing, the Board finds a challenged rate unreasonable, it will order the railroad to pay reparations to the complainant for past movements, 49 U.S.C. 11704(b), and may prescribe the maximum rate the carrier is permitted to charge for future movements, 49 U.S.C. 10704(a)(1). However, the Board may not set the maximum reasonable rate below the level at which the carrier would recover 180% of its variable costs of providing the service.[7]

In examining the reasonableness of a rate, the Board is guided by the multifaceted rail transportation policy set forth at 49 U.S.C. 10101. It must also give due consideration to the “Long-Cannon” factors contained in 49 U.S.C. 10701(d)(2)(A)-(C).[8] And the Board must recognize that rail carriers should have an opportunity to earn “adequate revenues.” 49 U.S.C. 10701(d)(2). Adequate revenues are defined as those that are sufficient—under honest, economical, and efficient management—to cover operating expenses, support prudent capital outlays, repay a reasonable debt level, raise needed equity capital, and otherwise attract and retain capital in amounts adequate to provide a sound rail transportation system. 49 U.S.C. 10704(a)(2).

In 1995, Congress added a new provision to the rail transportation policy calling for the “expeditious handling and resolution of all proceedings.” 49 U.S.C. 10101(15). It further instructed the Board to establish procedures to ensure expeditious handling of rail rate challenges in particular, including “appropriate measures for avoiding delay in the discovery and evidentiary phases of such proceedings.” 49 U.S.C. 10704(d). As previously stated, Congress also directed the Board to “establish a simplified and expedited method for determining the reasonableness of challenged rail rates in those cases in which a full stand-alone cost presentation is too costly, given the value of the case.” 49 U.S.C. 10701(d)(3).

Constrained Market Pricing

The Board’s general standards for judging the reasonableness of rail freight rates, which will continue to be applied to large rail rate disputes, are set forth in Guidelines. These guidelines adopt a set of pricing principles known as “constrained market pricing” (CMP). The objectives of CMP can be simply stated. A captive shipper should not be required to pay more than is necessary for the carrier involved to earn adequate revenues. Nor should it pay more than is necessary for efficient service. And a captive shipper should not bear the cost of any facilities or services from which it derives no benefit. Guidelines, 1 I.C.C.2d at 523-24.

CMP contains three main constraints on the extent to which a railroad may charge differentially higher rates on captive traffic.[9] The revenue adequacy constraint ensures that a captive shipper will “not be required to continue to pay differentially higher rates than other shippers when some or all of that differential is no longer necessary to ensure a financially sound carrier capable of meeting its current and future service needs.” Id. at 535-36. The management efficiency constraint protects captive shippers from paying for avoidable inefficiencies (whether short-run or long-run) that are shown to increase a railroad’s revenue need to a point where the shipper’s rate is affected. Id. at 537-42. The stand-alone cost (SAC) constraint protects a captive shipper from bearing costs of inefficiencies or from cross-subsidizing other traffic by paying more than the revenue needed to replicate rail service to a select subset of the carrier’s traffic base. Id. at 542-46.

A SAC analysis seeks to determine whether a complainant is bearing costs resulting from inefficiencies or costs associated with facilities or services from which it derives no benefit. The SAC analysis does this by simulating the competitive rate that would exist in a “contestable market.” A contestable market is defined as one that is free from barriers to entry. The economic theory of contestable markets does not depend on a large number of competing firms in the marketplace to assure a competitive outcome. Id. at 528. In a contestable market, even a monopolist must offer competitive rates or lose its customers to a new entrant. Id. In other words, contestable markets have competitive characteristics which preclude monopoly pricing.

To simulate the competitive price that would result if the market for rail service were contestable, the costs and other limitations associated with entry barriers must be omitted from the SAC analysis. Id. at 529. This removes any advantages that the existing railroad would have over a new entrant that create the existing railroad’s monopoly power. A stand-alone railroad (SARR) is therefore hypothesized that could serve the traffic at issue if the rail industry were free of entry barriers. Under the SAC constraint, the rate at issue cannot be higher than what the SARR would need to charge to serve the complaining shipper while fully covering all of its costs, including a reasonable return on investment. This analysis produces a simulated competitive rate against which we judge the challenged rate. Id. at 542.

To make a SAC presentation, a shipper designs a SARR specifically tailored to serve an identified traffic group, using the optimum physical plant or rail system needed for that traffic. Using information on the types and amounts of traffic moving over the railroad’s rail system, the complainant selects a subset of that traffic (including its own traffic to which the challenged rate applies) that the SARR would serve.

Based on the traffic group to be served, the level of services to be provided, and the terrain to be traversed, a detailed operating plan must be developed for the SARR. Once an operating plan is developed that would accommodate the traffic group selected by the complainant, the SARR’s investment requirements and operating expense requirements (including such expenses as locomotive and car leasing, personnel, material and supplies, and administrative and overhead costs) must be estimated. The parties must provide appropriate documentation to support their estimates.

It is assumed that investments normally would be made prior to the start of service, that the SARR would continue to operate into the indefinite future, and that recovery of the investment costs would occur over the economic life of the assets. The Board’s SAC analysis, however, only examines a set period of time.[10] The analysis estimates the revenue requirements for the SARR based on the operating expenses that would be incurred over that period and the portion of capital costs that would need to be recovered during that period. A computerized discounted cash flow (DCF) model simulates how the SARR would likely recover its capital investments, taking into account inflation, Federal and state tax liabilities, and the need for a reasonable rate of return. The annual revenues required to recover the SARR’s capital costs (and taxes) are combined with the annual operating costs to calculate the SARR’s total annual revenue requirements.

The revenue requirements of the SARR are then compared to the revenues that the railroad is expected to earn from the traffic group. There is a presumption that the revenue contributions from non-issue traffic (that is, the traffic of non-complaining shippers) should be based on the revenues produced by the current rates. Traffic and rate level trends for the traffic group are forecast into the future to determine the future revenue contributions from that traffic.

The Board then compares the revenue requirements of the SARR against the total revenues to be generated by the traffic group over the SAC analysis period. Because the analysis period is lengthy, a present value analysis is used that takes into account the time value of money, netting the annual over-recovery and under-recovery as of a common point in time. If the present value of the revenues that would be generated by the traffic group is less than the present value of the SARR’s revenue requirements, then the complainant has failed to demonstrate that the challenged rate levels violate the SAC constraint.

If, on the other hand, the present value of the revenues from the traffic group exceeds the present value of the revenue requirements of the SARR, then the Board must decide what relief to provide to the complainant by allocating the revenue requirements of the SARR among the traffic group and over time.[11]

Simplified Guidelines Adopted in 1996

Under Simplified Guidelines, the reasonableness of a challenged rate is to be determined by examining that challenged rate in relation to three benchmark figures. Each benchmark is expressed as a ratio of revenues to variable costs of providing rail service. The revenue-to-variable cost ratio is referred to as an R/VC ratio.

The first benchmark is called the Revenue Shortfall Allocation Method (RSAM). As currently designed, this benchmark measures the average markup over variable cost that the railroad would need to charge all of its “potentially captive” traffic in order for the railroad to earn adequate revenues as measured by the Board under 49 U.S.C. 10704(a)(2). Potentially captive traffic is defined as all traffic priced above the 180% R/VC level—which is the statutory floor for regulatory rail rate intervention.[12] Simplified Guidelines provided for the calculation and publication of an RSAM range. The upper end of the range reflects the average markup above variable cost that the railroad would need if it were to replace all of its assets as they wear out. The lower end subtracts out any shortfall related to movements priced below the 100% R/VC level. The lower end is an attempt to capture managerial inefficiencies. In Simplified Guidelines, however, the Board recognized that an R/VC ratio below 100% does not necessarily reflect improper pricing or a money-losing service. 1 S.T.B. at 1028-29. The precise RSAM benchmark the agency would use was therefore left unresolved, but was expected to fall within this range. Id. at 1029.

The second benchmark is called R/VC>180. As currently designed, this benchmark measures the average markup over variable cost earned by the defendant railroad on its potentially captive traffic. It could be more narrowly tailored to focus on a subset of the railroad’s traffic that has transportation characteristics similar to the traffic moving under the challenged rate.

The third benchmark is called R/VCCOMP. This benchmark is used to compare the markup being paid by the challenged traffic to the average markup assessed on other potentially captive traffic involving the same or a similar commodity moving similar distances.

The Board publishes tables each year showing the most recent RSAM range and the most recent R/VC>180 ratio for each Class I railroad, as well as regional averages. The R/VCCOMP ratios for appropriate comparison traffic were intended to be computed after a shipper files a rate complaint, using traffic data from the rail industry Waybill Sample,[13] and applying URCS costing.

The Board described these three benchmarks as “the starting point for a rate reasonableness analysis, not the end result.” Simplified Guidelines, 1 S.T.B. at 1022. The Board anticipated that both the shipper and railroad would present “whatever additional information is available that bears on the reasonableness of the pricing of the traffic at issue.” Id. The agency expressed confidence that careful analysis of these three benchmarks, together with whatever supplementary evidence is provided in a case, should enable the agency “to make at least a rough determination as to rate reasonableness in those cases where a more precise determination is not possible.” Id. at 1041.

Simplified Guidelines was challenged in court by the Association of American Railroads (AAR). The court dismissed the appeal, however, without ruling on the lawfulness of those guidelines, because the court concluded that it would benefit from seeing how the guidelines would be applied in a case. See Association of Am. R.R. v. STB, 146 F.3d 942 (D.C. Cir. 1998).

As previously stated, the Board held public hearings in April 2003 and July 2004 to examine why Simplified Guidelines has not been used by shippers and to explore ways to improve those guidelines. Based on these hearings, the Board concluded that revisions to the existing simplified guidelines appeared warranted. On July 28, 2006, the Board put out a detailed proposal for public comments. Initial comments were submitted in October 2006, reply comments in November 2006, and rebuttal comments in January 2007. A hearing was held on January 31, 2007, and final comments were received in February 2007.

Comments in this proceeding were submitted by: Alliance for Rail Competition (ARC) and PPL Energy Plus, LLC (PPL) (collectively ARC/PPL), American Short Line and Regional Railroad Association (ASLRRA), Arkansas Electric Cooperative Corporation (AECC), Association of American Railroads (AAR), BASF Corporation (BASF), BNSF Railway Company (BNSF), Canadian National Railway Company (CN), Canadian Pacific Railway Company (CP), Cargill, Inc. (Cargill), CF Industries, Inc. (CF Industries), Chevron Phillips Chemical Company LP (Chevron Phillips), Dow Chemical Company, LLC (Dow Chemical), E. I. Du Pont De Nemours And Company (Du Pont), National Industrial Transportation League (NITL), Norfolk Southern Railway Company (NS) and CSX Transportation, Inc. (CSXT) (collectively, NS/CSXT), Occidental Chemical Corporation (OCC), Olin Chemicals (Olin), Paducah & Louisville Railway, Inc. (P&L), Snavely King Majoros O’Connor & Lee, Inc. (SKMOL), Terra Industries, Inc. (Terra Industries), Union Pacific Railroad Company (UP), Kansas City Southern Railway Company (KCS), United Transportation Union – General Committee of Adjustment (UTU/GO-386), U.S. Clay Producers Traffic Association, Inc. (Clay Producers), U.S. Department of Agriculture (USDA), U.S. Department of Transportation (USDOT), United States Steel Corporation (US Steel), 11 wheat and barley commissions (Wheat & Barley Commissions), and six entities from the state of North Dakota (North Dakota).[14] A large group of shippers filed comments as a consolidated party (Interested Parties)[15] and a large group of agricultural shippers filed comments jointly (Agricultural Shippers).[16] After considering all of the comments, we are refining and adopting that modified proposal here.

Revised simplified guidelines

Our revised simplified guidelines are presented in three parts. Section I sets forth the Simplified-SAC method, designed to assess the reasonableness of challenged rates in medium-size rail rate disputes. Section II sets forth the refined Three-Benchmark method, intended for smaller rate disputes. Section III describes the procedural schedules and limits on discovery for both simplified methods. Section IV explains the limitations on the relief available under each of these simplified methods.

I. Simplified-SAC Methodology

CMP, with its SAC constraint, is the most accurate procedure available for determining the reasonableness of rail rates where there is an absence of effective competition.[17] The SAC test, which judges the reasonableness of a challenged rate by comparison to the rate that would prevail in a competitive market, rests on a sound economic foundation and has been affirmed by the courts. Under the SAC test, rate relief is available only where a captive shipper demonstrates that it is cross-subsidizing other parts of the defendant’s rail network or is bearing the costs of a carrier’s inefficiencies.

Any simplified methodology for assessing the reasonableness of rail rates should be designed to achieve the same objective, albeit in a less precise manner. In 1996, AAR advocated that the Board use a simplified SAC approach,[18] but at that time the Board was unable to design a feasible way to simplify the SAC test sufficiently to be cost-effective for smaller rail rate disputes. Our experience and expertise with SAC cases has grown substantially since then to the point where it is now possible to craft a Simplified-SAC approach. In the following discussion, we describe the objective and mechanics of the Simplified-SAC approach.

1. Objective

The principal objective of the SAC constraint is to restrain a railroad from exploiting market power over a captive shipper by charging more than it needs to earn a reasonable return on the replacement cost of the infrastructure used to serve that shipper. A second objective of the SAC constraint is to detect and eliminate the costs of inefficiencies in a carrier’s investments or operations.

It is the second objective that turns Full-SAC presentations into an intricate, expensive undertaking. To replicate less than the existing rail infrastructure used to serve the captive shipper, the complainant must demonstrate that there would still be sufficient capacity to handle expected demand. This requires the complainant to first select an appropriate subset of the railroad’s traffic for the SARR to serve, then design an operating plan that shows how an efficient railroad would serve this traffic group, and determine the optimal network configuration. Parties use complex computer programs to simulate the hypothetical SARR and test the operating plan and configuration against the forecasted traffic group. All these tasks are interrelated, such that changes to the traffic group may require reconfiguring the hypothetical network and revising the operating plan. The parties must then develop detailed evidence to calculate both the direct operating expenses (such as the costs of locomotives, crew, and railcars) and the indirect operating expenses (such as general and administrative and maintenance-of-way). The time and expense associated with this inquiry dwarfs the time and expense needed to examine the replacement cost of the necessary rail infrastructure.

Accordingly, the inquiry under the Simplified-SAC method described below will be limited to whether the captive shipper is being forced to cross-subsidize other parts of the railroad’s rail network. Such an approach will be a less precise application of CMP, because it will not identify inefficiencies in the current rail operation. But it will allow us to determine whether a captive shipper is being forced to cross-subsidize parts of the defendant’s existing rail network the shipper does not use.

To keep this critical inquiry as simple as possible, we will assume that all existing infrastructure along the predominant route used to haul the complaint traffic is needed to serve the traffic moving over that route. This is a reasonable simplifying approach. We recognize that in 1996 the Board rejected a different simplified SAC method that would have required a complainant to replicate the existing infrastructure. Simplified Guidelines, 1 S.T.B. at 1015. But rail capacity and traffic conditions have changed. Railroads no longer are burdened by substantial excess capacity; rather, the rail industry now faces the opposite situation. Rail capacity is strained, demand for transportation service is forecast to increase, and railroads must make capital investments to meet that demand.

Moreover, while a Simplified-SAC method may not fully implement CMP principles, see id., it more closely tracks CMP than does the Three-Benchmark approach. The Simplified-SAC method we propose would assure that a railroad does not earn unreasonable profits on its investments. As railroads enjoy increasing market power with rising demand for their services, the SAC test (in either its full or simplified form) will provide a critical restraint on their pricing of captive traffic, without deterring railroads from making the investments in their rail networks that are needed to meet rising demand. Indeed, the Simplified-SAC method would incorporate those new capital investments and ensure that the maximum lawful rate includes a reasonable return on the replacement cost of those investments.

Using this Simplified-SAC method should not create incentives for railroads to make inefficient investments. Our regulatory authority over rail rates is limited and competition controls the rates for most of a railroad’s traffic. Thus, railroads will have little incentive to deliberately gold-plate their rail infrastructure or make inefficient investments to influence the returns and rates this agency would permit under a Simplified-SAC constraint.[19] Rather, competition will force railroads to make prudent capital investments to meet forecast increases in demand for transportation services. And even if the management of some railroads is not as efficient as possible, the burden of uncovering and quantifying existing inefficiencies is so substantial as to be impracticable in all but the largest rail rate disputes. In any event, under this Simplified-SAC method, described in detail below, there will be some limited opportunities for a complainant to eliminate some costs associated with inefficiencies. A complainant could argue that some existing facilities (e.g., track, sidings, yards) along the selected route have fallen into disuse and should not be included.

2. Methodology

The Simplified-SAC presentation will differ from a Full-SAC presentation by eliminating or restricting the evidence parties can submit on certain issues. The core analysis in a Simplified-SAC proceeding will address the replacement cost of the existing facilities used to serve the captive shipper and the return on investment a hypothetical SARR would require to replicate those facilities. We will then seek to determine whether the traffic using those facilities is paying more than needed to cover operating expenses and a reasonable return on the replacement value of those facilities.

To hold down the cost of a Simplified-SAC presentation, various simplifying assumptions and standardization measures are essential. Towards that end, we will impose the following structure on a Simplified-SAC presentation:

         Route: The analysis will examine the predominant route of the issue movements during the prior 12 months for the traffic at issue.

         Configuration: The facilities of the SARR will consist of the existing facilities along the analyzed route (including all track, sidings, and yards). If a shipper presents compelling evidence that some facilities along the route have fallen into disuse by the railroad, and thus need not be replicated, those facilities will be excluded from the SAC analysis.

         Test Year: The Simplified-SAC analysis will examine the reasonableness of the challenged rates based on a 1-year analysis. The Test Year would be the most recently completed 4 quarters preceding the filing of the complaint.

         Traffic Group: The traffic group will consist of all movements that traveled over the selected route in the Test Year. No rerouting of traffic will be permitted.

         Cross-Over Traffic: The revenue from cross-over traffic will be apportioned between the on-SARR and off-SARR portions of the movement based on the revenue allocation methodology used in Full-SAC proceedings.[20]

         Road Property Investment: The Board’s findings in prior Full-SAC cases will be used to simplify parts of the road property investment (RPI) analysis. A more detailed discussion of how the RPI inquiry will be simplified is set forth in Appendix A.

         Operating Expenses: The total operating and equipment expenses of the SARR will be estimated using URCS. This will avoid the substantial debates over the operating plans and network configurations that consume much of a Full-SAC analysis. A more detailed discussion is set forth in Appendix B.

         Discounted Cash Flow Analysis: The DCF analysis will calculate the capital requirements of a SARR in the customary fashion and then compare the revenues earned by the defendant railroad against the revenue requirements of the SARR only for the Test Year.

         Internal Cross-Subsidy Inquiry: The internal cross-subsidy test set forth in PPL,[21] as refined in Otter Tail,[22] will be an affirmative defense, with the evidentiary burden of production and persuasion on the railroad.

         Maximum Reasonable Rate: The SAC costs (i.e., the revenue requirements of the SARR) will be allocated amongst the traffic group based on the methodology used in Full-SAC cases.[23]

         5-Year Rate Relief: The maximum lawful rate will be expressed as a ratio of revenue to variable costs, with variable costs calculated using unadjusted URCS. This maximum R/VC ratio would then be prescribed for a maximum 5-year period.

II. Three-Benchmark Methodology

For some shippers who have smaller disputes with a carrier, even this Simplified-SAC method would be too expensive, given the smaller value of their cases. These shippers must also have an avenue to pursue relief. Accordingly, we will retain the Three-Benchmark method for those shippers, with refinements to lessen the uncertainties of the existing method.

We will adopt the following changes to the method described in Simplified Guidelines:

         Waybill Sample: provide both parties access to the unmasked Waybill sample of the defendant carrier(s), subject to customary protective orders, upon the filing of a complaint;

         Variable Cost Calculations: use only unadjusted URCS to calculate the variable cost of the issue movement and all movements in the comparison group;

         Non-Defendant Traffic: exclude non-defendant traffic from the comparison group;

         R/VCCOMP: use a final-offer procedure to select the comparison group most similar in the aggregate to the challenged movement;

         RSAM and R/VC>180: use an unadjusted RSAM figure and revise the way these benchmarks are calculated;

         Rate Reasonableness Determination: adjust each movement in the comparison group by the ratio of RSAM R/VC>180, calculate a “confidence interval” around the estimate of the mean of the adjusted comparison group, and presume unreasonable a challenged rate that is above this confidence interval;[24] and

         Other Relevant Factors: permit either the shipper or the carrier to rebut the presumption with evidence of “other relevant factors.”

1. Comparison Group

a. Comparability Factors

The purpose of the R/VCCOMP benchmark is to use the R/VC ratios of other “potentially captive traffic” (i.e., traffic priced above the 180% R/VC level) as evidence of the reasonable R/VC levels for traffic of that sort. As such, the comparison group should consist of only captive traffic over which the carrier has market power. The rates available to traffic with competitive alternatives would provide little evidence on the degree of permissible demand-based differential pricing needed to provide a reasonable return on the investment. Moreover, we are comparing mark-ups over variable cost to determine the reasonable level of contribution to joint and common costs for a particular movement. This means that movements with different cost characteristics may be included in the comparison group. For example, if a complainant challenged the rate for a 6-car to 25-car movement, it may argue for the inclusion of a comparable movement of a 50-car to 110-car unit train by another potentially captive shipper, or vice versa. While the rate associated with the larger unit-train movement will be lower to reflect greater efficiencies, there is no reason, a priori, to presume that the R/VC ratios (or their share of joint and common costs) should be different. However, because we are using URCS to develop the variable costs for the issue movement and comparison movements, we will favor a comparison group that consists of movements of like commodities so the variable cost calculation of the issue movement and comparison group will be similar.

Accordingly, comparability will be determined by reviewing a variety of factors, such as length of movement, commodity type, traffic densities of the likely routes involved, and demand elasticity (although the comparison group need not have movements with identical demand). The selection of the best comparison group will be governed by which group the Board concludes provides the best evidence as to the reasonable level of contribution to joint and common costs for the issue movement.

b. Selection Process

The selection process will begin with the shipper and railroad simultaneously tendering their initial evidence regarding an appropriate comparison group. The movements must be drawn from the Waybill Sample provided to the parties by the Board at the outset of the case. Shortly after receipt of the initial tenders, designated Board staff may convene a technical conference with the parties to discuss and attempt to resolve any disputes as to the appropriateness of movements in the comparison groups.

Each party must then tender its “final offer” group of movements it believes should comprise the comparison group. Only movements that had previously been submitted by one of the parties in its initial tender can be included in the final offer groups. In other words, each party can select its final comparison group only from movements contained either in its first tender or in the first tender of the other side. Any movement set forth in both sides’ initial tenders will be required to be included in each side’s final comparison group, unless the parties agreed to exclude the movement. After the submission of the final offer comparison groups, each party will be given an opportunity to challenge the other party’s comparison group and support its own in simultaneous rebuttal filings.

The Board will then select the comparison group that it concludes is most similar in the aggregate to the issue movements. This will be an “either/or” selection, with no modifications by the Board. A final offer procedure for determining the comparison group is in the public interest because it will encourage both parties to submit a reasonable comparison group. Any final tender that is skewed too far in one direction might well result in the selection of a more reasonable final tender presented by the opposing party. By having two rounds of simultaneous tenders and a technical conference, both sides will participate in the winnowing process. Each side therefore should be able to provide a reasonable final offer comparison group, even if the two sides’ groups differ. Thus, the Board will only have to determine which group is more reasonable. This should enable a prompt, expedited resolution of the comparison group selection. This approach will work as intended only if the parties know that the agency will not attempt to find a compromise position somewhere in the middle.[25] To create the proper incentives for the litigants not to take extreme positions, we commit to selecting the more reasonable of the two groups as tendered.

2. RSAM Range

The RSAM benchmark is intended to measure the average markup above variable cost that the carrier would need to charge to meet its own revenue needs. However, when Simplified Guidelines were adopted in 1996, the Board did not settle on a single formula for computing this benchmark. The Board explained that it “[did] not believe that the industry ha[d] yet become so efficiently sized that all of its current assets were used and useful and would warrant replacement as they wear out,” and it suggested that the necessary revenue contribution was therefore less than what would be needed to provide for the replacement of all existing assets. Simplified Guidelines, 1 S.T.B. at 1029. Accordingly, the Board decided to look at the effect on a carrier’s revenue needs of subtracting out any shortfall related to movements priced below the 100% R/VC level, which the Board referred to as a “managerial efficiency adjustment,” even though the Board acknowledged that an R/VC ratio below 100% does not necessarily reflect improper pricing or a money-losing service. Id. at 1028. The end result was publication of an RSAM range that would form “the relevant starting range for [the Board’s] consideration.” Id. at 1030. The RSAM benchmark the agency would use in a particular case was left unresolved, but was expected to fall within this range.

In the NPRM, we proposed to eliminate the range and use the unadjusted RSAM in the rate comparison approach. As no party opposed that proposal, we will adopt it for the reasons set forth in the NPRM.

3. Method To Calculate RSAM and R/VC>180

We are changing the way RSAM is calculated to address a flaw in the existing method. RSAM has been calculated by computing the uniform mark-up above variable cost that would be needed from all potentially captive traffic “for the carrier to recover all of its URCS fixed costs.” Simplified Guidelines, 1 S.T.B. at 1027.[26] When a carrier is not “revenue adequate” under the Board’s annual calculations, its RSAM figure (what it needs to collect) should be greater than its R/VC>180 figure (what it is actually collecting). Conversely, when a carrier is revenue adequate under that determination, its RSAM figure should be lower than its R/VC>180 figure.

But this relationship between RSAM and R/VC>180 has not held true in the RSAM calculations. For example, the 2002 RSAM and R/VC>180 figures show that the unadjusted RSAM figure for the Norfolk Southern Railway Company (NS) (216%) was less than its R/VC>180 figure (221%), suggesting that NS was revenue adequate.[27] Yet NS was not revenue adequate in that year.[28] The opposite, erroneous relationship between RSAM and R/VC>180 can be seen in the most recent calculations, where the relationship between RSAM and R/VC>180 would indicate that NS was revenue inadequate in 2004, even though NS in fact earned the target rate of return (the railroad industry’s average cost of capital) that year.[29]

To address this concern, we are altering slightly the way RSAM and R/VC>180 are calculated. The R/VC>180 benchmark will be derived from the confidential Waybill Sample. We will use that sample to estimate the total revenue earned by the carrier on potentially captive traffic (REV>180) and the total variable costs of the railroad to handle that traffic (VC>180). We will not adjust these calculations to match the total revenue and cost information reported by the carriers in their annual filings to the Board, as has been our prior practice, as the entities may differ and we conclude the adjustments complicate the analysis without improving the accuracy of the estimate. Accordingly, R/VC>180 will be calculated as follows:

R/VC>180 = REV>180 VC>180

To calculate RSAM, we will add to the numerator the carrier’s revenue shortfall (or subtract any overage) shown in our annual revenue adequacy determination (REVshort/overage). RSAM will then be calculated as follows:

RSAM = (REV>180 + REVshort/overage) VC>180

Recalculated in this manner, the ratio of RSAM to R/VC>180 will reflect how far the railroad is over or under the revenue adequacy target.

Table 1 shows the impact of these changes on the relationship between the two benchmarks for a sample year (2004). In a rate case, we will not rely on the figures from a single year, but will use a 4-year average where possible.

Table 1

Benchmark Comparison (2004)

Railroad

Current Approach

(with range)

 

New Approach

(as recalculated)

RSAM

(1)

R/VC>180

(2)

Ratio

(3)=(1)(2)

 

RSAM

(4)

R/VC>180

(5)

Ratio

(6)=(4)(5)

BNSF

215 – 266

234

0.92 - 1.14

 

305

234

1.31

CSXT

254 – 292

197

1.29 - 1.48

 

297

228

1.30

GTC

322 – 375

233

1.38 - 1.61

 

404

253

1.60

KCS

241 - 298

259

0.93 - 1.15

 

300

263

1.14

NS

197 - 226

212

0.93 - 1.07

 

228

243

0.94

SOO

234 - 331

261

0.90 - 1.27

 

353

244

1.45

UP

245 - 306

210

1.17 - 1.46

 

324

232

1.40

 

4. Rate Reasonableness Presumption & Other Relevant Factors

Once the Board finds that the railroad has market dominance over the movements at issue, the Board will select the appropriate comparison group through the final-tender process described above. Each movement in the comparison group would then be adjusted by the ratio of RSAM R/VC>180. The Board will then calculate the mean and standard deviation of the R/VC ratios for the adjusted comparison group (weighted in accordance with the proper sampling factors).

If the challenged rate is above a reasonable confidence interval around the estimate of the mean for the adjusted comparison group, it will be presumed unreasonable and, absent any “other relevant factors,” the maximum lawful rate will be prescribed at that boundary level. Using the mean (R/VCCOMP) and standard deviation (S) of the adjusted comparison group, along with the number of movements in the comparison group (n), the upper boundary of a reasonable confidence interval around the estimate of the mean would be derived as follows:[30]

upper boundary = R/VCCOMP + tn-1 (S (n-1))

This confidence interval would be a function of the number of movements in the comparison group and the standard deviation of those adjusted R/VC ratios. A small standard deviation or large number of observations would produce a tighter confidence interval, so that we could have more “confidence” in the accuracy of our estimate of the mean of the comparison group.

Parties may submit evidence of “other relevant factors” to demonstrate that the maximum lawful rate should be higher or lower. Parties are required, however, to quantify the impact of these “other relevant factors” on the presumed maximum lawful rate. For example, a shipper could not submit evidence that the railroads are not operating as efficiently as possible without quantifying the extent of the inefficiency and how that should affect the presumed maximum lawful rate.

To keeps these cases manageable, we must impose certain limits on the nature of the “other relevant factors” evidence we will consider and the breath of discovery we will permit. Evidence of product and geographic evidence associated with particular movements will not be permitted, nor will we permit evidence of movement-specific adjustments to URCS. We reserve the right to prohibit other categories of evidence if experience demonstrates that the introduction of such evidence would or does unduly complicate this process, which must be relatively simple and inexpensive to have any value. Similarly, in reviewing discovery disputes by either party over evidence of other relevant factors, we will scrutinize the burden placed on the party from which discovery is sought. Even if information requests are clearly relevant, if the burden is considerable we might not require the discovery.

III. Procedural Matters

1. Procedural Schedules

We establish a tight procedural schedule for small rate disputes, using designated Board staff to assist in resolving discovery disputes and to chair technical conferences. The schedules are set forth below. If a deadline falls on a weekend or holiday, that deadline will be extended until the next business workday, but the remaining deadlines would remain unchanged. The Board will consider deviations from this procedural schedule only upon a good cause showing by the party. However, for a Simplified-SAC case, if there is a dispute between the parties over the predominant route of the issue movements during the Test Year, the Board will stay the close of discovery and the deadline for the railroad’s second request until that issue is resolved.

 

 

 

 

 

Table 2

Procedural Schedule for Three-Benchmark Case[31]

 

Complaint

Day 0

Complainant Initial Disclosure

Day 0

STB production of unmasked Waybill Sample

Day 10

Mandatory Mediation Begins

Day 10

Answer to complaint

Day 20

Railroad Initial Disclosure

Day 20

Mediation Period Ends

Day 30

Discovery Commences

Day 30

Discovery Closes

Day 60

Opening (Complainant)

Day 90

Opening (Railroad)

Technical Conference

Day 95

Reply (Complainant)

Day 120

Reply (Railroad)

Rebuttal (Complainant)

Day 150

Rebuttal (Railroad)

Board Decision Within 90 Days

 

Table 3

Procedural Schedule for Simplified-SAC Case

Complaint

Day 0

Complainant Initial Disclosure

Day 0

Mediation Begins

Day 10

Answer to Complaint

Day 20

Railroad Initial Disclosure

Day 20

Mediation Concludes

Day 30

Discovery Begins

Day 30

Railroad Second Disclosure

Day 140

Discovery Closes

Day 150

Opening Evidence

Day 220

Reply Evidence

Day 280

Rebuttal Evidence

Day 310

Technical Conference

Day 320

Final Briefs

Day 330

Board Decision Within 180 Days

 

The procedural schedules provide for a mandatory 20-day period of non-binding mediation. Board staff will be appointed to mediate these disputes. To protect the confidentiality of mediation discussions, the appointed Board staff will be recused from all subsequent involvement in the case should the case not be fully resolved through mediation. The entire mediation process will be confidential, including all material used or exchanged and positions taken by the parties. The mediation period can be extended at the consent of the parties. Designated representatives from the parties with authority to settle the dispute shall participate in all meetings, unless the Board-appointed mediator concludes such involvement is not necessary. In a Three-Benchmark case, the Board will release the confidential Waybill Sample, subject to the proper protective orders, before the mediation begins, to facilitate settlement.

If the traffic at issue is part of a class of traffic that has been exempted from Board regulation pursuant to 49 U.S.C. 10502(a), the complainant will need to file a separate request for revocation of the pertinent class exemption for the traffic at issue pursuant to 49 U.S.C. 10502(d). The Board will generally consider the revocation request before permitting a rate challenge. If a revocation request is accompanied by a complaint, the procedural schedule set forth above (including mediation) will generally be stayed automatically pending the outcome of the request for revocation.

2. Discovery

To streamline the discovery process, certain standardized discovery will be required to be produced with the complaint and answer, and technical conferences will be held to resolve factual disputes early. These and other matters related to discovery are set forth below.

Staff Conferences. As has been our practice in Full-SAC cases, we will use designated Board staff to facilitate voluntary resolution of discovery disputes and to conduct technical conferences.

Meet and Confer Requirement. Parties will be required to meet and confer on discovery and procedural matters within 7 business days after the mediation period ends. As soon as possible, the parties must inform the Board whether there are unresolved disputes that require Board intervention and, if so, the nature of those disputes.

Complainant’s Initial Disclosures. The complainant will be required to provide certain initial disclosures concurrent with the filing of its complaint. At that time, the complainant will provide to the railroad its preliminary estimate of the variable cost of the challenged movements, using the unadjusted figures produced by the URCS Phase III program, demonstrating that the Board’s jurisdictional threshold has been met.[32] The complainant will also provide to the railroad all documents that it relied upon to determine the inputs to the URCS Phase III program. In addition, the complainant shall include with its complaint a narrative addressing whether there is any feasible transportation alternative for the challenged movements, and disclose to the railroad all documents relied upon in formulating that assessment.

Railroad’s Initial Disclosure. The railroad will likewise be required to provide initial disclosures to the complainant concurrent with filing its answer. Like the shipper, the railroad shall produce its preliminary estimate of the variable cost of each challenged movement, using the unadjusted figures produced by the URCS Phase III program. And the railroad must provide to the complainant all documents that it relied upon to determine the inputs used in the URCS Phase III program.

Railroad’s Second Disclosure in Simplified-SAC Cases. In Simplified-SAC cases, the railroad shall provide the following additional information to the complainant:

         Identification of all traffic that moved over the routes replicated by the SARR in the Test Year;

         Information about those movements, in electronic format, aggregated by origin-destination pair and shipper, showing the origin, destination, volume, and total revenues from each movement;

         Total operating and equipment cost calculations for each of those movements, computed in accordance with Appendix B, and provided in electronic format, so the complainant can readily estimate the total operating and equipment costs of the SARR;

         Revenue allocation for the on-SARR portion of each cross-over movement in the traffic group, developed in accordance with the methodology used in Full-SAC cases, provided in electronic format;

         Total trackage rights payments paid or received during the Test Year associated with the route replicated by the SARR;

         All workpapers and documentation necessary to support these calculations.

Discovery and Interrogatory Requests in Three-Benchmark Cases. In Three-Benchmark cases, we will limit the number of discovery requests that either party can submit to the other party without obtaining advance authorization from the Board. Each party would be limited to ten interrogatories (including subparts), ten document requests (including subparts), and one deposition.

Motions to Compel. Motions to compel will be governed by 49 CFR 1114.31 (a)(2)‑(4). Any appeals to the Chairman of a ruling by Board staff on a motion to compel will be due within 3 business days of the ruling. Replies to the appeal will be due within 3 business days after the appeal is filed. Criteria set forth in 49 CFR 1115.9(a) will govern the standard of review for such interlocutory appeals.

3. Jurisdictional Inquiry

The Board may investigate the reasonableness of a challenged rate only where the revenues the carrier receives for transporting the movements at issue exceed 180% of its variable costs of providing the service. This jurisdictional threshold for rail rate regulation also serves as the floor for regulatory relief, because the Board cannot prescribe a rate below the jurisdictional threshold.[33] By statute, a carrier’s variable costs are to be determined using URCS with adjustment only where the Board finds it appropriate.[34]

The Board will use its unadjusted URCS model to determine the variable costs for a rail carrier. If the carrier is not a Class I carrier, the Board will use the most appropriate regional URCS data. The only adjustments allowed to the URCS Phase III program would be those adopted in Ex Parte No. 431 (Sub-No 2). See Review of the General Purpose Costing System, 2 S.T.B. 754 (1997); Review of the General Purpose Costing System, 2 S.T.B. 659 (1997). Those adjustments include the so-called “270” volume shipment adjustments, the make-whole adjustments, TOFC/COFC adjustments, and RoadRailer adjustments. In addition, the circuity factor is always set to one when actual miles are used to calculate the variable costs.

IV. Eligibility Criteria

While crafting the basic structure of the simplified approaches has been challenging, creating eligibility limitations on what cases should be decided under each approach has been especially difficult. Captive shippers must have an effective forum to bring rail rate disputes. But an overly simplified approach should not be applied to a case when the amount in dispute justifies the use of a more robust and precise approach. Seeking the right balance is complicated by the lack of historical data on the cost of presenting evidence under a Simplified-SAC or Three-Benchmark approach. A more profound complication flows from asking the Board to determine the likely “value” of a dispute at the outset of a case without prejudging the merits of the case.

In an effort to avoid prejudging the merits of a case, the Board proposed an eligibility criteria based on a more objective criterion: the Maximum Value of the Case (MVC). The MVC of a complaint would have been the maximum relief the shipper could attain over 5 years if the challenged rates were reduced to the jurisdictional floor (i.e., the level at which the R/VC ratio equals 180%). The MVC would have been calculated by multiplying the difference between the challenged rate and the rate floor by the annual volume of the traffic at issue. If a complaint challenged multiple rates covering different origins and destinations, the Board would have aggregated the MVC for each set of movements covered by the complaint. In this fashion, the MVC would have been equal to the net present value, as of the time of the filing of the complaint, of the maximum relief that the shipper could obtain.

Based on the record compiled in this proceeding, it is clear that the MVC proposal is inadequate to the task. It was our intent to carve out certain cases that could plainly qualify to use either the Simplified-SAC or Three-Benchmark approach based on objective criteria, and leave cases on the margin to be analyzed on a case-by-case basis. But public comments demonstrated that few if any movements would actually qualify for simplified treatment under that approach.[35] Creating a safe harbor based on the maximum value of the case would leave most captive shippers in the same situation they have been in for 10 years: having to undertake a potentially expensive and uncertain process of proving their right to use the simplified methods at the outset of the case.

In commenting on our initial MVC proposal, the carriers struck on an alternative approach—under which a complainant could stipulate to a limit on relief sought in order to use a simplified approach—that we believe achieves the right balance. Accordingly, we adopt a “limit to relief” approach, under which a complainant may elect to use either the Simplified-SAC or Three-Benchmark approach, but will be limited in the relief available under those approaches. Section 1 below describes this approach in detail. Section 2 addresses the carriers’ reversal of position and opposition to their own refinement. Section 3 sets forth our cost estimates to present a Full-SAC, Simplified-SAC, and Three-Benchmark case. Section 4 then describes how the approach incorporates the “risk factor” sought by the shipper community. Section 5 discusses our decision to remove the “aggregation” proposal and Section 6 discusses the potential for carriers to improperly force shippers to use a more expensive approach.

1. Relief Limits

Complainants that proceed under the Three-Benchmark methodology will be limited to $1 million of rate relief over a 5-year period, and complainants that elect to proceed under the Simplified-SAC methodology will be limited to $5 million dollars of relief over the same period.[36] Each limit is based on our estimates of the litigation cost to pursue relief under the next more complicated, and more precise method. By permitting a shipper to use either of these simplified approaches, we ensure that the rate complaint process will be available for any size rate dispute. But by placing limits on the relief available, we encourage shippers with larger disputes to pursue relief under the more appropriate methodology without the Board itself trying to determine the likely value of a case. Instead, the complainant must evaluate its own claim, decide for itself the expected value of the case, and balance the value against the litigation costs and the potential relief it may receive.

Shippers have expressed concern that a captive shipper should not be required to commit to a particular approach before it has an opportunity for discovery that might lead it to revise its assessment of the value of the case. We will address this concern by permitting a complainant to amend its complaint and seek relief under a different methodology at any time prior to the filing of opening evidence. If a complainant wishes to amend its complaint to pursue relief under a more complex approach, it must pay the relevant filing fee and the procedural schedule would begin anew, except we would not again require the parties to submit to mandatory mediation. However, to protect carriers from potentially endless amendments to the complaint, the shipper will have an automatic right to amend its complaint without prejudice only once.

The limit on relief will apply to the difference between the challenged rate and the maximum lawful rate, whether in the form of reparations, a rate prescription, or a combination of the two. Any rate prescription will automatically terminate once the complainant has exhausted the relief available. Thus, the actual length of the prescription may be less than 5 years if the shipper ships a large enough volume of traffic so that the relief is used up in a shorter time. The complainant will be barred from bringing another complaint against the same rate for the remainder of the 5-year period.

Once a rate prescription expires, the carrier’s rate making freedom will be restored with a regulatory safe harbor at the challenged rate for the remainder of the 5-year period, with appropriate adjustments for inflation using the rail cost adjustment factor, adjusted for inflation and productivity (RCAF-A).[37] If, however, a carrier establishes a new common carrier rate once the rate prescription expires, and the new rate exceeds the inflation-adjusted challenged rate, the shipper may bring a new complaint against the newly established common carrier rate. In this way, the shipper will be discouraged from using a cruder methodology than the value of the case warrants, but a railroad does not get a potentially massive regulatory windfall from an exhausted prescription.

We believe that these limits on relief strike the appropriate balance. Captive shippers with disputes of any size can now avail themselves of our rate review processes, while carriers can be assured that a large rate dispute will not be subjected to a more simplified process than necessary to achieve that objective.

Text Box: Value of the Case

As illustrated above, shippers with disputes up to $1 million over 5 years may use the Three-Benchmark approach, our simplest rate reasonableness methodology. Any shipper who believes the value of its dispute exceeds $1 million should pursue a Simplified-SAC case. Likewise any shipper that believes the value of its case exceeds $5 million should offer a Full-SAC presentation. There may be instances where a complainant will be faced with a difficult choice between forgoing some of the potential value of a dispute or pursuing greater relief despite increased costs. But we conclude that it is appropriate to encourage litigants facing this choice to use the rate reasonableness approach that is best suited for the magnitude of the dispute.

By adopting clear lines of demarcation for eligibility, we will meet our stated goal of providing clear, usable guidance as to which complainants may use the simplified methods. This limit of relief approach should expedite cases by avoiding protracted disputes over eligibility. Further, bright line demarcations provide regulatory certainty that should foster negotiation.

2. Carrier Objections

The Board’s original proposal was to place limits on the availability of these simplified procedures based on the maximum value of the dispute. Because the calculation of the maximum value depended on the traffic volume of the disputed movement, we proposed that shippers would stipulate to the amount of traffic it intended to ship. Recognizing that this approach would overstate the actual value of the case, the carriers suggested a further stipulation: that shippers could lower the maximum value of the case (and thus qualify for a more simplified approach) by stipulating to the minimum rate that they would seek. This proposal was endorsed by all the Class I carriers without any reservations.[38] Under the carriers’ proposed refinement, any shipper would be able to use the simplified procedures by stipulating to modest traffic volume, a little rate reduction, or a combination of the two.

Shippers were intrigued by the carriers’ proposal and sought an extension of the procedural deadline to consider the refinement.[39] Ultimately, they could not support the proposal over concerns that a shipper would need to stipulate to a limit on relief without sufficient information to make an informed choice.

In advance of our hearing, we asked the parties to comment on a modest simplification of the carriers’ approach that would address the shippers’ concerns.[40] Rather than have the shippers stipulate to both the maximum volume and minimum rate to use a simplified process, the Board would establish a limit on the total relief available. To address shipper concerns over the lack of information needed to make an informed choice, the Board proposed to permit the shipper to amend its complaint anytime before the filing of opening evidence, and to opt into a more robust (or more simplified) approach based on its reassessment of the value of its case.

Following our hearing, however, the carriers reversed course and opposed what was essentially their own proposal. The AAR stated that it “could not endorse this approach in the abstract. Whether it could be acceptable to AAR depends at a minimum on the eligibility criteria adopted by the Board, which is precisely the issue that remains unsettled. Moreover, AAR has concerns that in application the small claims model might have nothing to do with small cases and everything to do with tactics.”[41] UP stated that it was not its intent to permit any shipper to use a simplified approach.[42]

The rules adopted here are very similar to the railroads’ refinement to the original proposal. The fact that the carriers now oppose the approach, apparently not having foreseen the practical consequences of their own proposal, is difficult to understand or give considerable weight. Moreover, the railroads’ concerns are unpersuasive. Rather, we believe that the selected limits on relief we are placing on each simplified approach will provide an effective safeguard against misuse of either simplified approach. They will also achieve the central objective that any captive shipper—regardless of its size or the amount it ships—will be able to pursue a level of relief appropriate to its case.

3. Litigation Cost Estimates

Based on the record and our experience with Full-SAC litigation, we will base the limits on relief on our estimate that it should cost the shipper no more than $5 million to present a Full-SAC case and $1 million to conduct a Simplified-SAC analysis.

Counsel for Otter Tail testified that it cost that company $4.5 million to pursue its Full-SAC case.[43] While many of the comments received confirm our belief that the changes recently adopted in Major Issues will serve to lower these litigation expenses, the extent of the cost reductions remains to be seen. Moreover, we cannot ignore the fact that Full-SAC litigation has been very costly and time consuming in recent years and the cost and complexity have increased over time. We cannot predict what new or unforeseen issues will emerge. But we are confident that the costs to present a Full-SAC case should be no more than the $5 million. Thus, we will use $5 million as the limit on relief available under the Simplified-SAC approach. If the reforms enacted in Major Issues have the intended effect and Full-SAC litigation costs decline, we can revisit the $5 million limit at a later date.

The litigation cost of a Simplified-SAC presentation is more difficult to discern, but should be dramatically less than the cost of presenting a Full-SAC case. We carefully simplified and streamlined the SAC process to create the Simplified-SAC process, and the refinements to that process adopted—removing rerouting of traffic and the annual true-up—should simplify the analysis and reduce the costs further.

Interested Parties submitted a line-item estimate of the cost to prepare a Simplified-SAC complaint prepared by a transportation consultant.[44] No other party has offered detailed evidence, though some critiqued the presentation by Interested Parties.[45] In the absence of any better evidence, we are guided by Interested Parties’ estimate with the following modifications. Interested Parties’ testimony was premised on the Simplified-SAC approach described in the NPRM. As we have made further simplifying modifications to the proposal here, we have adjusted the witness’s estimate by deducting those cost elements that are no longer part of the Simplified-SAC analysis. Further, where we conclude the estimates are overstated, we have reduced the number of consultant hours to a more reasonable amount.[46] As a result, we conclude that a Simplified-SAC presentation will cost approximately $1 million in consulting and legal fees. Thus, we will adopt $1 million as the limit to relief for Three-Benchmark cases.

We conducted a similar analysis, also set forth in Appendix C, to derive a cost estimate of $250,000 to present a case under the Three-Benchmark approach. While this figure is also drawn from the shipper’s testimony, we believe this estimate is very conservative and reflects the most that a Three-Benchmark case should cost to litigate. It includes a substantial increase in litigation cost to cover the possibility that a shipper may seek to submit “other relevant factors.” Without such evidence, which is optional on the shipper’s part, the litigation estimate is below $200,000, and should be reduced further once a body of precedent is developed to guide the implementation of the Three-Benchmark approach.

These cost estimates are based on the best evidence of record. As cases are brought under these two approaches, actual litigation costs will become available, and over time we will observe whether the reforms of Major Issues have had the desired effect of reducing litigation costs in Full-SAC cases. At that time, parties may petition the Board to adjust the limit on relief as needed, assuming they provide detailed litigation cost estimates.

4. Risk Factor

Shippers note that when any litigant considers pursing a legal claim, the potential relief must exceed the costs to litigate the case by a fair margin to justify the expense. They refer to this cushion or margin between the costs and the potential relief as a “risk factor.” While conceding that it would be difficult to develop a precise estimate of the proper risk factor, they ask that in setting limits on relief, we incorporate a risk factor of at least three.

We agree with the general principle that a simplified presentation would not be cost-effective unless the potential relief exceeds the expected cost of obtaining the remedy by a sufficient margin to make it worthwhile to pursue the complaint. Accord Simplified Guidelines, 1. S.T.B. at 1049. The limit to relief approach should make it cost effective to pursue either a Three-Benchmark of Simplified-SAC presentation by providing an ample cushion between the cost to bring the case and the potential relief available. Table 4 sets forth the expected litigation costs, potential relief, and risk factor associated with the three approaches.

Table 4

 

Expected Cost to Litigate

Potential Relief Available

Risk Factor

Three Benchmark

$250,000

$1 million

4

Simplified-SAC

$1 million

$5 million

5

Full-SAC

$5 million

no limit

n.a.

 

The shipper community would like the potential relief available to dramatically exceed these limits. But their “risk factor” analysis is flawed as it approaches the question from the wrong direction. The error flows from comparing the relief available under one approach to the cost of bringing a different approach. They would establish the limit on relief under the Three-Benchmark approach by taking the cost to litigate the Simplified-SAC case and multiplying that cost by 3. Under this approach, the limit on relief would be set at $15 million and $3 million for Simplified-SAC and Three-Benchmark approaches, respectively. But the potential relief available under the Three-Benchmark approach would exceed 12 times the cost to litigate the case, and the risk factor for a Simplified-SAC presentation would exceed 15. As such, while shippers start with the desire to create a risk factor of 3, their approach results in risk factors several times higher.

5. Aggregation of Claims

The limits on relief that we establish here do not include a mechanical mechanism to police against attempts to divide a large dispute into multiple smaller disputes. It is not clear that such a mechanism is necessary at this time. The Board has ample discretion to protect the integrity of its processes from abuse, and we should be able to readily detect and remedy improper attempts by a shipper to disaggregate a large claim into a number of smaller claims, as the shipper must bring these numerous smaller cases to the Board.

6. Carrier Manipulation of Dispute

At least one shipper raised concerns that any eligibility approach premised on the challenged rate may permit the carrier to force the shipper into a more expensive methodology simply by raising the challenged rate. So, for example, if a small shipper wanted to challenge a rate under the Three-Benchmark approach, but the carrier facing litigation elected to raise the rate so the annual transportation charges increased by $200,000, the small shipper would have little choice but to bring a Simplified-SAC case.

We acknowledge this potential and will carefully review any allegations of carrier misconduct on a case-by-case basis. Should we conclude that the carrier has improperly sought to force the shipper to use a more expensive methodology by raising the challenged rate, we will either remove the limits on relief entirely for that case, or increase the limit on rate relief to the next threshold level, depending on the equities of the case.

Regulatory Flexibility Act

P&L argues that the Board must engage in a more detailed Regulatory Flexibility Act analysis explaining and providing support for the Board’s conclusion that the proposed rules would not adversely impact small businesses. P&L maintains that the proposed guidelines could have a major economic impact on many small railroads, but that the Board has not weighed the costs that the proposed standards would impose on shortline and regional railroads. P&L argues that the proposed guidelines do not take into account the increased administrative costs that smaller railroads would incur in evaluating and revising internal costing and ratemaking systems to comply with the proposed standards. Additionally, P&L claims that carriers similarly situated to P&L would have to familiarize themselves with URCS and a regulatory scheme that, according to P&L, essentially imposes a Class I costing system on smaller railroads.[47]

P&L’s arguments are without merit. The URCS analyses contemplated here do not require any additional record keeping by Class II or Class III carriers and do not require any reporting of any additional information to the Board. Moreover, local movements on Class II and Class III carriers are excluded from the Three-Benchmark method, reducing small carrier’s exposure to these rules. Under the adopted rules, some small carriers would also have an opportunity to avoid litigation in conjunction with a Class I rate by rebilling their own separate rates. And the majority of railroads and shippers likely to be involved in proceedings under these Simplified Guidelines are unlikely to be small entities within the meaning of the Regulatory Flexibility Act. Finally, these carriers have been subject to the Simplified Guidelines since 1996, and while our decision to revise and reform those guidelines in this decision may change the approach in certain respects, it does not expand the universe of traffic subject to our regulation. Accordingly, we certify that this action will not have a substantial adverse impact upon a significant number of small entities.

This action will not significantly affect either the quality of the human environment or the conservation of energy resources.

CONCLUSION

These new simplified rail rate guidelines address many of the concerns raised about Simplified Guidelines. They provide shippers meaningful access to regulatory relief in those cases where a Full-SAC case is too costly, given the value of the case. They also promote the rail transportation policy to protect captive shippers from unreasonable rates, 49 U.S.C. 10101, without precluding rail carriers from earning revenues that are adequate under honest, economical, and efficient management, 49 U.S.C. 10704(a)(2).

Conflicting federal policies guided our analysis. We must balance the shippers’ interest in being protected from unreasonable rates, see 49 U.S.C. 10101(6), against the need to promote a safe and efficient rail transportation system by allowing rail carriers to earn adequate revenues, see 49 U.S.C. 10101(3); 49 U.S.C. 10701(d)(2). Moreover, Congress specifically directed the agency to create a simplified and expedited method for determining the reasonableness of challenged rail rates in those cases where a Full-SAC presentation is too expensive, given the value of the case. 49 U.S.C. 10701(d)(3). Our analysis of the Waybill Sample, shown in Table 5, indicates that, Full-SAC presentation would be impractical for 73% of potentially captive traffic. It also shows that, even a Simplified-SAC presentation would be too costly for 45% of potentially captive traffic.[48] These simplified procedures will provide a meaningful forum for the resolution of rail rate disputes arising out of the at least 73% of traffic that previously was prevented from bringing rate complaints to the Board due to the high costs of developing a Full-SAC presentation.

Table 5

Revised Eligibility Estimate

Description

All Regulated
R/VC>180

Full-SAC

Simplified-SAC

3-Benchmark

Farm Products

$942,701

$99,111

11%

$279,741

30%

$563,849

60%

Metallic Ore

199,375

117,913

59%

57,705

29%

23,758

12%

Coal

3,168,171

1,663,985

53%

901,532

28%

602,654

19%

Crude Petroleum

2,403

-

0%

766

32%

1,637

68%

Non Met. Minerals

111,508

16,134

14%

39,382

35%

55,991

50%

Ordnance

7,267

-

0%

5,148

71%

2,119

29%

Food Products

280,429

15,990

6%

60,724

22%

203,715

73%

Chemicals

2,534,098

170,500

7%

649,539

26%

1,714,058

68%

Petroleum Products

516,563

76,150

15%

122,518

24%

317,894

62%

Rubber Products

43

-

0%

-

0%

43

100%

Stone, Clay, Glass

353,647

26,919

8%

134,423

38%

192,305

54%

Mach., Excl Elec.

6,311

-

0%

5,773

91%

538

9%

Mach., Elec.

12,534

5,280

42%

3,034

24%

4,220

34%

Transp. Equip.

23,158

9,326

40%

4,373

19%

9,458

41%

Waste or Scrap

40,906

5,884

14%

21,141

52%

13,881

34%

Misc. Frt.

103,195

28,259

27%

56,608

55%

18,328

18%

Misc. Mix

7,667

4,309

56%

398

5%

2,960

39%

Small Pkg Frt.

2,375

-

0%

1,965

83%

410

17%

Hazardous Wastes

20,895

-

0%

5,732

27%

15,163

73%

 

8,333,245

2,239,759

27%

2,350,502

28%

3,742,984

45%

 

We have resolved these conflicting policies by creating the Simplified-SAC process, retaining the Three-Benchmark approach, and placing reasonable limits on the relief available under both simplified approaches. We conclude this strikes a reasonable balance between providing a simplified method that permits captive shippers to seek protections from unreasonable rates, while encouraging use of the most precise approach feasible for the amount in dispute.

As the guidelines are tested through application to actual rail rate disputes, we will garner greater understanding of the litigation costs of these approaches and whether the proposed structure is working. And the creation of a body of precedent will provide guidance to the rail community on some of the remaining ambiguities in the approach. Nonetheless, we intend to carefully monitor the application of these guidelines, and remain vigilant that the goals of simplification are not thwarted as parties begin to litigate cases under these guidelines. While these new guidelines reflect an important step forward in creating a workable structure for resolving rate disputes of all sizes, we anticipate that further steps will be needed as the application of the guidelines reveal unanticipated issues or show that more or less simplification is warranted.

Changes to the Code of Federal Regulations needed to implement this proposal are set forth in Appendix D and will be published in the Federal Register.

It is ordered:

 

1.      The simplified rail rate guidelines discussed above are adopted.

 

2.      This decision is effective on October 7, 2007.

 

By the Board, Chairman Nottingham, Vice Chairman Buttrey, and Commissioner Mulvey. Vice Chairman Buttrey concurring with a separate expression.

 

Vernon A. Williams

Secretary

 

 

____________________________________

 

VICE CHAIRMAN BUTTREY, concurring:

The simplified rail rate case guidelines that we adopt here represent another step in this agency’s continuing effort to make meaningful rail rate dispute resolution accessible to small shippers in a timely and cost-effective manner.

 

In 1887, the genesis of U.S. railroad regulation arose from concerns about rates for grain shippers. Unfortunately, this concern is still with us. Based on the Board’s own analysis and that of the U.S. Government Accountability Office (GAO) Report of October, 2006, the changes that have occurred in the rail industry since the Staggers Rail Act of 1980 have been largely positive, except for certain grain rates. According to the GAO report, the amount of grain traffic with comparatively high markups over variable cost continues to increase. The Board held a hearing on Rail Transportation of Grain on November 2, 2006, at which the concerns of grain shippers were heard. The U.S. Congress has also expressed concern about the availability of meaningful rail rate dispute resolution for small grain shippers.

 

The simplified guidelines we adopt today will make the Three-Benchmark methodology available for all small rail rate disputes including those involving small grain shippers. A low filing fee ($150), mandatory mediation, fast-track schedule, early access to waybill sample information, use of unadjusted URCS data and the availability of substantial rate relief should all make this methodology attractive. In the all-important selection of the comparison traffic group (the traffic to which the challenged movement will be compared), our decision makes clear that a small shipper can include traffic of other potentially-captive shippers of the same commodity, even if that traffic moves in larger blocks. In other words, a 6-car grain shipper can include 50-car or 110-car unit-train traffic in its comparison group. As the decision explains, the Three-Benchmark methodology will compare the mark-up over variable cost to determine the reasonable level of contribution to joint and common costs for a particular movement. While the rates associated with the larger unit-train movement should, on average, be lower to reflect greater efficiencies, it is the R/VC ratios, not the rates, that will be compared in the Three-Benchmark methodology.

 

Small grain shippers, and small shippers generally, need some assurance that our processes will give them a more reasonable opportunity to prevail in a small rate case if their rates are unusually high. I join in support of this decision because I believe that it can offer that assurance. I hope that small shippers will avail themselves of the new methodology, and I will be vigilant to make sure that the new rules are fairly applied in these cases.


Appendix A – Road Property investment

The RPI component of a Full-SAC analysis has remained fairly consistent in recent cases, even though the average investment includes a mix of heavy- and light-density lines, as well as varying yards along the route. Segregating recent decisions regionally results in a very consistent set of RPI costs (indexed to 2005). For example, as shown below, the average RPI cost per track mile has varied less than 10% in the last five western Full-SAC cases.

Table A-1

Western SAC Case RPI Cost (2005 Dollars)

 

Total RPI

($ Millions)

Track

Miles

Cost per

Track Mile

Otter Tail

$2,865

1,563

$1,883,146

Xcel

1,396

678

2,058,877

TMPA

4,850

2,403

2,018,220

PPL

618

296

2,086,374

WPL

3,713

1,765

2,103,695

 

However, differences in geography and configuration can have a significant impact on the investment costs, as seen by comparing the RPI figures of the western Full-SAC cases to those of the eastern Full-SAC cases. Thus, using an aggregate investment cost per mile could sacrifice too much accuracy for simplicity.

Instead, we will use elements from prior cases that do not change significantly from case-to-case (such as the unit costs of earthwork), and adjust the Simplified-SAC RPI analysis for those features that will vary significantly from case-to-case (such as the number of bridges, miles of track, and acres of land). The parties will account for inflation by applying an appropriate index to the findings from prior cases. (The numbers reported below have not been indexed, except for the consolidated bridge trend curves.) We will use a rolling average from past cases, such that as new Full-SAC cases are issued by the Board, older cases will be dropped from the comparison in subsequent Simplified-SAC proceedings. We will not include decisions prior to TMPA, nor decisions that did not resolve all of the disputes between the parties, such as in PPL.

RPI is broken down into the following categories: (1) land, (2) roadbed preparation, (3) track, (4) tunnels, (5) bridges and culverts, (6) signals and communication, (7) buildings and facilities, (8) public improvements, (9) mobilization, (10) engineering, and (11) contingencies. See, e.g, Otter Tail at E1-E6. We will simplify the calculation of each of these categories as set forth below.

1. Land

We will use a rolling-average cost per acre from prior rate cases. Table A‑2 below shows the Board’s land cost per acre findings, by category of land:[49]

Table A-2

Comparison of Per Acre Land Costs by Category

 

Year[50]

Agricultural

Residential

Industrial

Commercial

Otter Tail

2002

$533

$13,006

$14,844

$32,423

Duke/NS

2002

4,088

3,853

76,611

204,849

Duke/CSXT

2002

4,141

6,982

39,842

94,656

CP&L

2002

3,932

4,913

83,253

130,900

Xcel

2001

446

22,157

13,797

42,549

TMPA

2001

4,932

24,709

47,234

74,344

 

We recognize that land prices are affected by location, but the costs to present individualized valuation evidence outweigh the benefits. The cost for a narrow corridor of land that is in virtually all cases predominately in rural areas is a very small part of the total RPI. Thus, even a large change in land cost per acre will not have a significant impact on the overall analysis. Moreover, developing land valuation evidence is expensive, as parties must hire an appraiser to conduct real estate appraisals for land across thousands of miles. This typically includes paying the expert to survey the entire right-of-way (ROW). On balance, we believe that, for a Simplified-SAC presentation, simplifying this component of the RPI analysis is warranted, as the added precision does not justify the high costs of developing more accurate land valuations.

2. Roadbed Preparation

Roadbed preparation is a significant category of RPI that is less subject to simplification on a route-mile basis than the others, as it can be affected by both the terrain and makeup of the route being replicated. For example, there are significant economies in preparing roadbed for double track rather than single track. And roadbed preparation for a rail line over a mountain is much more costly than roadbed preparation for a rail line through the flat American heartland.

Major simplifications remain possible, however. The parties can and should continue to use the ICC Engineering Reports to determine the underlying quantities of material needed for line segments where these data have been reported. They should convert the ICC-reported quantities to current engineering standards using the methodology currently in use in Full-SAC cases. (The Board will make available sample spreadsheets from prior Full-SAC cases for parties to use upon request.) For line segments for which there are no ICC data, the parties must present evidence on the quantities of material needed under current engineering standards. Following current Board precedent in Full-SAC cases, we will assume that ditches should be 2 feet by 2 feet in size, that the right-of-way (ROW) will be 100 feet across, that adequate access roads are reflected in the current quantities, and that side slopes will be 1.5 to 1.

The unit costs for earthwork—by far the largest component of roadbed preparation—will be based on the rolling average from past Full-SAC cases. The Board has been consistent in the mix of required equipment to perform roadbed preparation. These costs can be expressed in unit cost per cubic yard of material for excavation, loose rock, solid rock, borrow, and in some cases, fine grading. Table A-3 below shows the Board’s unit cost findings for roadbed preparation from prior Full-SAC cases. (Fine grading costs per unit are available only for the Otter Tail and Xcel cases.)

Table A-3

Comparison of Earthwork Unit Costs (per cubic yard)

 

Common

Loose

Solid

Borrow

Fine Grading

Otter Tail

$3.90

$6.57

$9.22

$12.35

$0.33

Duke/NS

3.32

8.75

9.09

9.84

 

Duke/CSXT

3.29

8.67

9.09

9.81

 

CP&L

3.34

8.81

9.20

9.89

 

Xcel

3.43

8.00

9.57

12.26

0.15 slope
0.32 subgrade

TMPA

3.19

4.51

7.15

10.46

 

 

The remaining miscellaneous earthwork costs (such as seeding and topsoil) will be estimated on a route-mile basis. While there is some variation in this expense category between cases, the total cost of these miscellaneous earthwork costs is a relatively minor part of the overall RPI analysis, so that more precise estimates would have only a modest, if any, impact on the SAC analysis. Table A-4 below shows the Board’s findings from prior Full-SAC cases.

Table A-4

Comparison of Other Earthwork Unit Costs

 

Total Cost

($ Millions)

Route

Miles

Cost per

Route Mile

Otter Tail

$43.8

1,208

$36,260

Duke/NS

91.6

1,108

82,643

Duke/CSXT

93.8

1,197

78,399

CP&L

79.1

818

96,555

Xcel

21.7

367

59,027

TMPA

54.3

1,629

33,303

 

3. Track

We will use the rolling average track cost per track mile (excluding ballast and subballast) from prior rate cases.[51] Parties are expected to submit unit costs on a track mile basis for ballast and subballast including transportation costs. Table A-5 below shows the Board’s prior findings regarding total costs per track mile, excluding ballast and subballast.

 

Table A-5

 

Comparison of Track Construction Costs

 

 

Total Cost
($ Millions)

Track
Miles[52]

Cost per
Track Mile

Otter Tail

$744.5

1,563

$476,342

Duke/NS

$693.9

1,382

$502,087

Duke/CSXT

$712.4

1,510

$471,816

CP&L

$508.3

1,073

$473,693

Xcel

$358.1

678

$528,123

TMPA

$1,271.2

2,403

$528,999

 

 

 

 

Note: Ballast and subballast costs excluded

4. Tunnels

We will not simplify this part of the RPI analysis. There have been only a few Full-SAC cases dealing with the cost of tunnels, and those costs are specific to each individual tunnel. Thus, if there is a tunnel on the ROW replicated by the SARR, the parties must submit evidence on the current replacement cost of that tunnel.

5. Bridges and Culverts

There are two acceptable methods the parties may use to calculate bridge costs. The first method calculates actual bridge costs based on unit costs from prior cases for specific bridge types. The second method bases costs on a representative sample of the average cost at two foot intervals to build bridges up to 350 feet in length.

The first method uses a cumulative average bridge cost per linear foot from prior rate cases.[53] As all bridges are not the same, we will use the bridge cost for the appropriate type of bridge. The parties will need to submit evidence on the total length (by type) of the bridges along the ROW being replicated, but then use the rolling-average unit costs from prior cases. For Eastern bridges, Type 1 bridges consist of all bridges with lengths from 10 to 40 feet; Type 2 bridges range from 41 to 75 feet; and Type 3 bridges have lengths above 75 feet. For Western bridges, Type 1 bridges are pre-stressed concrete girder bridges; Type 2 bridges are steel deck plate girder bridges; and Type 3 bridges are steel through plate girder bridges. We will assume that inclusion of assets in the ICC Engineering Reports is adequate proof of bridge ownership. Tables A-6 and A-7 show the Board’s prior bridge cost per linear foot findings, by type of bridge.[54]

Table A-6

Comparison of Eastern Bridge Construction Costs

 

Type 1

Cost per foot

Type 2

Cost per foot

Type 3

Cost per foot

Duke/NS

$6,044

$3,405

$3,813

Duke/CSXT

4,892

3,924

3,993

CP&L

5,790

3,967

3,701

 

Table A-7

Comparison of Western Bridge Construction Costs

 

Type 1

Cost per foot

Type 2

Cost per foot

Type 3

Cost per foot

Otter Tail

$2,315

$2,552

$4,300

Xcel

1,793

2,690

4,427

TMPA

2,225

3,862

4,409

 

In addition to the method described above, parties may estimate bridge costs for bridges sharing local terrain characteristics of the Western bridges based on bridge length without consideration of bridge design or height.[55] The parties must submit the total length of each bridge along the ROW. Bridge costs for bridges greater than 350 feet in length cannot use this alternative method due to the lack of significant data points for longer bridges. Table A‑8 below shows the bridge cost trend curves from the Board’s prior western SAC cases. Future cases will be cumulatively added to our current data points.

Table A-8

 

STB Derived Trend Curves for Western Bridges

 

 

y=$/linear foot, x=length

Western SAC Cases

y= -.0295x3+15.824x2+1659.4x+24917

 

 

Culvert costs will be estimated using the rolling average culvert cost per linear foot from prior rate cases. As all culverts are not the same, we will use the culvert cost for the type of culvert involved: corrugated metal pipe (CMP), reinforced concrete box culvert (RCB), and structural steel plate pipe (SSP). As each of these types of culverts are utilized on railroads in a plethora of different sizes, most of the culvert evidence that has been submitted in previous Full-SAC cases includes a linear equation that correlates the cross-sectional area of the culvert opening with the unit cost of the culvert. We will utilize these linear regressions to determine the cost per foot of all the various sizes of culverts in our simplified analysis. The parties must submit evidence on the total length (by type) of culverts along the ROW being replicated, and then use the rolling-average unit cost (from the regression equations) from prior cases. Table A‑9 below shows the Board’s findings on the regression equations for culvert cost per linear foot by type of culvert.

Table A-9

Comparison of Culvert Construction Costs (per LF)

 

CMP Culvert

RCB Culvert

SSP Culvert

y=$/LF

x=sq in

y=$/LF

x=sf

y=$/LF

x=sq in

Otter Tail

y=0.0392x+17.606

y=4.017x+172.3

y=0.0171x+72.524

Duke v. NS

y=0.0277x+8.89

y=8.681x+134.609

y=0.0162x+145.59

Duke v. CSX

y=0.0276x+8.89

y=8.671x+134.295

y=0.0161x+145.66

CPL v. NS

y=0.025x+11.322

y=4.563x+198.47

y=0.0161x+163.875

Xcel

y=0.0304x+26.399

y=3.886x+286.052

y=0.00934x+155.158

TMPA

y=0.0237x+14.695

y=3.726x+266.77

y=0.0127x+145.201

 

For example, if the SARR in a Simplified-SAC presentation would replicate 2,000 feet of CMP culverts with a diameter of 10 square inches, the parties would first use the six equations above to calculate the CMP culvert construction cost per linear foot. They would then index those unit costs by the appropriate index. We would then use the rolling average of those (indexed) costs per linear foot, multiplied by 2,000 feet, to derive the culvert construction costs for the SARR. The following chart shows the data consistency.

Chart A-1

Sample Culvert Unit Costs

6. Signals and Communication

The overall cost of signals and communication have been consistent across cases. However, complainants in Full-SAC cases have constructed SARRs where the majority of lines would have centralized traffic control (CTC) signalling. Taking that signalling cost per mile and applying it to the entire ROW replicated by the SARR in a Simplified-SAC proceeding could overstate this expense category when the railroad does not itself use CTC signalling along that ROW.

Therefore, where CTC signalling is used by the railroad along the ROW replicated by the SARR, parties will use the rolling average signal and communications cost per route mile from prior rate cases. If, however, the railroad instead uses an automatic block system (ABS) along the route at issue, the complainant can submit evidence on the replacement cost of the facilities needed for that signalling technology. If a complainant is satisfied that the costs per mile from our prior findings are a suitable surrogate or will not be material to the outcome, it may elect to use the rolling-average cost per mile from prior Full-SAC cases for the entire SARR. Complainants electing not to use the rolling average costs for signalling will also have to present evidence on the communications costs. Table A-10 below shows our findings on signalling and communications costs per route mile.

Table A-10

Comparison of Signalling & Communications Costs (with CTC)

 

Total Cost

($ Millions)

Route

Miles

Cost per

Route Mile

Otter Tail

$203.8

1,208

$168,669

Duke/NS

154.8

1,108

139,689

Duke/CSXT

187.8

1,197

156,914

CP&L

138.7

818

169,578

Xcel

76.8

367

209,142

TMPA

133.4

1,629

81,883

 

7. Buildings and Facilities

Upon review of the cost of buildings and facilities per ton (total traffic volume), we observed a trend in the data. Chart 2 below reveals noticeable economies associated with the cost of buildings and facilities.

Chart A-2

Buildings & Facilities Costs

While the cost of buildings and facilities increases with more traffic (reflecting the larger workforce of the SARR), the cost per ton falls because of economies of scale. As such, doubling the size of the SARR does not double the size and cost of buildings and facilities needed to support the staff.

Accordingly, rather than using a rolling average from past cases, the parties must estimate the relationship between cost per ton and tonnage using a simple regression analysis of the costs from prior rate cases. Once this relationship is estimated using the data from the most recent cases, the parties will use this estimated relationship, combined with the total tons flowing over the SARR in the Test Year, to develop the buildings and facilities costs. Table A‑11 below shows the Board’s findings regarding buildings and facilities cost per ton (in the base year of those SAC presentations), which the parties will use to perform the regression analysis.

Table A-11

Comparison of Building & Facilities Costs

 

Total Cost

($ Millions)

Forecast Volume

(Millions)

Cost per

Ton

Otter Tail

$51.3

219.6

$0.234

Duke/NS

39.0

77.9

0.500

Duke/CSXT

62.0

104.9

0.591

CP&L

37.9

72.3

0.524

Xcel

41.2

105.3

0.391

TMPA

53.2

178.6

0.298

 

8. Public Improvements

We will use the rolling-average public improvement cost per route mile from prior Full-SAC rate cases. Public improvements without grade separations are the smallest cost category within RPI—averaging approximately $25,000 per route mile. The large disparity between cases is primarily due to fencing costs, as well as geographic differences between eastern and western cases. Although there is a variance from case-to-case, in the Board’s most recent Full-SAC decisions, estimates of these unit costs were very close to the overall average.

Grade separations, however, are a large and location-specific cost item within public improvements. (A grade separation is where a rail line crosses a road using either an overpass or underpass.) Therefore, we will calculate the rolling average cost for public improvements (without grade separation costs) on a route mile basis and calculate a separate rolling average cost for grade separations, weighted by the number of separations. The Board has accepted 10% of the cost of constructing grade separations in past Full-SAC cases where the railroad shows it contributed some level of investment, and we will do so in a Simplified-SAC proceeding.

Tables A-12 and A-13 show the Board’s findings.

Table A-12

Comparison of Public Improvement Costs (Without Grade Separations)

 

Total Cost

($ Millions)

Route

Miles

Cost per

Route Mile

Otter Tail

$29.5

1,208

$24,391

Duke/NS

17.3

1,108

15,575

Duke/CSXT

3.7

1,197

3,549

CP&L

7.6

818

9,313

Xcel

12.3

367

33,597

TMPA

75.8

1,629

46,521

 

Table A-13

Comparison of Grade Separation Costs

 

Total Cost

($ Millions)

Number of

Separations

Cost per Separation

Otter Tail

$9.6

17

$561,877

Duke/NS

16.9

8

2,117,957*

Duke/CSXT

3.7

7.9

469,857

CP&L

3.3

6

554,317

Xcel

8.8

16.3

539,225

TMPA

23.3

28

832,437

* Given the outlier nature of this observation, it will be excluded from the rolling average.

 

9. Mobilization, Engineering, and Contingencies

Mobilization will be fixed at 3.5% of the cost of road preparation, track, tunnels, bridges and culverts, signals and communications, buildings and facilities, and public improvements. Engineering will be fixed at 10% of the same RPI expense categories. Contingencies will be fixed at 10% of road preparation, track, tunnels, bridges and culverts, signals and communications, buildings and facilities, public improvements, mobilization, and engineering. This follows our practice in Full-SAC proceedings.

 


Appendix B – Total SARR Operating Expenses

This appendix describes the calculation of total operating expenses and equipment costs for the SARR in a Simplified-SAC proceeding.[56]

I. Background

In a Full-SAC case, parties develop operating expenses based upon three components: the traffic group, network configuration, and operating plan. The traffic group determines the volume of traffic the SARR would need to carry, expressed in tons, and the associated revenues of that traffic. The network configuration defines the route miles, track miles, yards, joint facilities, grades, curves, etc. of the SARR network. The operating plan defines the types of service provided, train sizes, types of locomotives, types of railcars, train speeds, loading times, unloading times, interchange times, crew change locations and times, dwell times at yards, inspection locations, train control systems, and signal systems of the SARR.

In more recent Full-SAC cases, information regarding the traffic group, network configuration and operating plan have been input into a computer simulation model know as the Rail Traffic Control (RTC) model to develop transit times and other operating statistics. The RTC model simulates actual operations of the SARR—coordinating the meets and passes of trains much like dispatchers do operating a real railroad—to determine the feasibility of the operating plan and the resulting operating statistics. Those operating statistics are then used to develop operating expenses.

In Full-SAC cases, operating expenses are separately calculated for 15 different categories of costs.[57] Developing these operating expenses is a complex process, with parties filing volumes of evidence supporting their respective positions. Parties argue at great length over the inputs used to calculate operating expenses. The complexity adds greatly to the expense of bringing, defending and adjudicating rate cases. The rules adopted here are designed to dramatically simplify this analysis by using the Board’s general purpose costing model (URCS).

II. Operating Expenses in Simplified-SAC Case

In a Simplified-SAC case, the traffic group will consist of all traffic traveling over the segments of track of the defendant’s rail network actually used to serve the complaint movement. The configuration will be the actual configuration of that part of the defendant’s network. Therefore, the system-average operating expenses of the railroad should provide a reasonable surrogate for the operating expenses to handle this traffic group.

We can use system-average URCS costs because the traffic that would share the SARR facilities should reflect a mix of traffic with some combination of unit train traffic, intermodal traffic, and general merchandise traffic. Some traffic may use more than the average number of locomotives, for example, while other traffic may use fewer. But the combined operating characteristics of the traffic group should be close to the system average. Thus, the system-average operating costs should provide a reasonable surrogate for calculating the operating expenses for all of the traffic traveling over the SARR.

Significant adjustments to the URCS values are needed for this task, however. First, the values produced by URCS need to be modified to reflect total, rather than variable, operating expenses. Second, because RPI would be calculated separately, the values produced by URCS would need to be adjusted to exclude those costs. Third, because many costs would be addressed in the DCF analysis, the URCS values would need to be adjusted to remove those expenses. Finally, there are a few types of expenses that a railroad may incur that are not included in a Full-SAC analysis and likewise should be excluded from a Simplified-SAC analysis. These adjustments are outlined below.

There is one category of operating expenses used in Full-SAC cases—training and start-up costs—that will be understated by using URCS. Training costs are included in URCS only to the extent railroads have on-going training expenses. Start-up costs of the sort reflected in recent Full-SAC cases are not included in URCS because the Class I railroads, as existing entities, do not incur these costs on a yearly basis. In the public comments, no party provided a reasonable way to reflect start-up expenses in a Simplified-SAC analysis. Therefore, only actual training expenses incurred by the railroad will be included in a Simplified-SAC analysis. Because this cost component is only a small part of the operating expenses of a SARR in a Full-SAC case, the omission should not significantly affect the outcome of a Simplified-SAC analysis.

The following list describes how specific elements within URCS will be handled in Simplified-SAC cases:

1.      Accumulated Deferred Tax Credits. This URCS item will be excluded because the DCF accounts for the impact of taxes. Including accumulated deferred tax credits would be a double count.

2.      Construction Work In Progress - Account 90. This URCS item will be excluded because there will be no construction work in progress once the SARR is built.

3.      Dismantling Retired Road Property. This URCS item will be excluded because, over the time period of the SAC analysis, this expense would not be incurred by the SARR.

4.      Equipment Capital Cost. This URCS item will be included at the historic book value less depreciation for the reasons discussed in Appendix C below.

5.      Interest During Construction - Account 76. This URCS item will be excluded because interest during construction is included in the DCF model.

6.      Other Elements of Investment - Account 80. This URCS item will be excluded because there will be no applicable expense for the SARR.

7.      Road Property Investment. This URCS item will be excluded because it will be separately accounted for in the Simplified-SAC analysis.

8.      Roadway Machines - Property Account 37. This URCS expense item will be included because it is part of roadway maintenance and would not be reflected in the RPI analysis.

9.      Shop Machinery Investment - Property Account 44. This URCS expense item will be excluded because it will be included in the RPI analysis.

10.  Working Capital. This URCS item will be excluded because it is not included in Full-SAC cases.

11.  Trackage Rights Payments: This URCS item will be excluded and the defendant carrier will be required to produce during discovery the total trackage rights payments paid or received during the Test Year associated with the route.

12.  100% Flow Thru Mode. We will run the URCS program in the 100% Flow Thru mode for calculating operating and equipment costs.[58]

In the NPRM, we proposed to have Board staff perform all these adjustments to URCS by adjusting the data files used by URCS, and make these modifications available to the public. We conclude, however, that the Board would benefit from having the parties propose the necessary modification to the URCS files in individual cases. This will subject the modifications to the rigorous scrutiny of litigation, where any issues in the implementation of this approach can be fully aired before the Board endorses any particular approach. Once we have enough experience to settle upon a particular approach, we will consider whether to provide two versions of URCS: the normal URCS and a modified URCS for use in a Simplified-SAC proceeding.


Appendix C – Public Comments & Board ResponSeS

Our summary of, and response to, the public comments is divided into five parts. In Section I, we address comments to the Simplified-SAC approach. In Section II, we address comments to the Three-Benchmark approach. In Section III, we address comments to our reliance on unadjusted URCS to determine the jurisdictional threshold. In Section IV, we summarize the comments received on our eligibility proposal. And in Section V, we address various miscellaneous comments. All comments were reviewed and analyzed, even those not specifically referenced.

I. Simplified-SAC Approach

1. Congressional Intent

Several shippers claim that the proposed Simplified-SAC methodology does not comply with Congress’ directive in 49 U.S.C. 10701(d)(3) to develop simplified and expedited guidelines for the handling of small rate disputes. In this regard, the shippers make three arguments: that the statute does not permit a three-tiered approach; that the statute requires the Board to adopt a methodology that differs from SAC; and that the Simplified-SAC methodology is neither simplified nor expedited.

a. Three-Tiered Approach

In section 10701(d)(3), Congress directed that “[t]he Board shall, within one year after the effective date of this paragraph, complete the pending Interstate Commerce Commission non-coal rate guidelines proceeding to establish a simplified and expedited method for determining the reasonableness of challenged rail rates in those cases in which a full stand-alone cost presentation is too costly, given the value of the case.” Contrary to the shippers’ claim, nothing in that language prevents us from establishing separate methodologies for small- and medium-sized cases. The statute instructed the Board to establish a simplified “method,” and we have concluded that the best “method” is the creation of separate processes for rail rate disputes of varying size.

Moreover, the three-tiered approach represents sound regulatory policy. It would not be practical to offer only the Simplified-SAC approach, as there will be many cases where the cost and time of bringing a Simplified-SAC case would outweigh the value of the case. On the other hand, it would be inappropriate to judge all but the very largest disputes based on a method as crude as the Three-Benchmark method. We believe that it is the shippers’ one-or-the-other argument—which would require us to abandon this practical three-tiered system for one that is either prohibitively expensive in some cases or lacking in precision in other cases—that would be contrary to the goal of the statute.

b. Use of a SAC-Based Method

The shippers claim that the statute precludes the Board from adopting a methodology that is based on the SAC test. We disagree. The statute directed the Board to establish a simplified and expedited method for those cases in which a “full” SAC presentation is too costly. Congress’ use of the modifier word “full” indicates that a simplified, less expensive version of the SAC would be consistent with Congressional intent. Moreover, as noted in the NPRM (at 9), CMP, with its SAC constraint, provides the most accurate procedure available for determining the reasonableness of rail rates where there is an absence of effective competition because CMP rests on a sound economic foundation. We do not believe that Congress would have wanted the Board to reject a methodology with a solid economic foundation in favor of a cruder approach less well grounded in economic principles.[59]

ARC/PPL argues that Full-SAC is the method used in coal rate cases, and that by adopting a simplified version of SAC, the Board has failed to comply with Congress’ directive to develop “non-coal rate guidelines.”[60] While it is true that the majority of Full-SAC cases have involved rates for coal traffic, the Simplified-SAC approach will be cost-effective for rate cases involving a much wider range of commodities. That approach, when combined with the option of pursuing relief under the Three-Benchmark approach, satisfies Congress’ goal of creating rail rate guidelines suitable for non-coal cases.

c. Simplified and Expedited

Several shippers also argue that the Simplified-SAC method is neither a simplified nor expedited methodology and thus does not comply with Congress’ directive in section 10701(d)(3). They observe that the proposed procedural schedule for completion of a Simplified-SAC proceeding will be 545 days, which is longer than the 480 days schedule in 49 CFR 1118 and 49 U.S.C. 10704(a)(3) for completion of a Full-SAC case.[61]

The shippers’ claim is without merit, as it is clear that Simplified-SAC will significantly expedite resolution of rate complaints, as compared to the Full-SAC process. While our default schedule is 480 days for Full-SAC cases, the complexity of the Full-SAC analysis has delayed the procedural schedule in every Full-SAC case. In most instances, the delays in Full-SAC cases have been the result of extended discovery and problems with the evidentiary record. Indeed, it has been necessary for the Board to issue a “compliance order” in most recent Full-SAC cases to obtain supplemental evidence to avoid gaps in the record and mismatches in the parties’ evidence associated with the operating plans of the SARR. These delays will be avoided in a Simplified-SAC case, as we have fixed the traffic group and configuration of the SARR and eliminated the expensive and complex inquiry into the optimal operating plan for the SARR.

To support the argument that the Simplified-SAC procedures are not simplified, Interested Parties point to the details of proposed rules for this method laid out in the NPRM. However, the shippers make this point without providing the appropriate context; specifically, that Congress charged the Board with adopting procedures that are simplified in comparison to the Board’s Full-SAC methodology. See 49 U.S.C. 10701(d)(3). The Simplified-SAC approach is vastly simplified in comparison to the Full-SAC method. In particular, it simplifies the following complex and expensive disputes in every Full-SAC case: rerouting of traffic, selection of the traffic group, traffic revenue and volume forecasts, SARR’s configuration, RPI costs, and operating plan and expenses. Moreover, the procedures adopted here have been further simplified from what was originally proposed by eliminating the annual update of any rate prescription and eliminating rerouting of the issue movement.

2. Testing

Various shippers argue that we must test the Simplified-SAC methodology to determine if the reduced costs will exceed the possible higher rates that the method could produce, or whether the results of Simplified-SAC are comparable to those reached in a Full-SAC case.[62] In support of this claim, Interested Parties observe that the Board tested a Simplified-SAC procedure advanced by the AAR in 1993, as part of the original Simplified Guidelines proceeding.[63]

There is no requirement, either in the statute or in our precedent, that the agency test guidelines in advance. Neither the SAC test nor the original Three-Benchmark approach was tested in advance.[64] Nor is the fact that we tested the AAR’s simplified SAC proposal in original rulemaking instructive. That model was a black-box, computer model offered by AAR. Because it was a proprietary computer program that came from an outside party, and the program was not open for either the Board or other parties to examine, the Board tested the computer model to see if the assumptions embedded in the program were valid (and the testing suggested the assumptions were invalid). Here, the Simplified-SAC proposal did not come from an outside party, but from the Board itself.

More fundamentally, it is impossible for the Board to test the Simplified-SAC method. To conduct such testing, we would have to undertake both a Full-SAC and Simplified-SAC analysis for the same traffic. And, as the Interested Parties observed, we cannot test these guidelines on past coal cases, but must instead examine non-coal traffic that would more likely use the Simplified-SAC procedures.[65] Even if we had the information for such an inquiry—which we do not, nor is it clear we can compel the railroads to produce that information in the absence of a complaint—it would be a tremendously expensive endeavor given the scope and level of detail required for a Full-SAC analysis. We do not have the budget or manpower to conduct a Full-SAC and Simplified-SAC analysis for even a single hypothetical movement. Yet the use of averages and URCS in the Simplified-SAC approach virtually guarantees that the results of the Full-SAC will differ in some respect from that of a Simplified-SAC. Therefore, to have any probative value, the testing would need to be conducted more than once (for different commodities, routes, and carriers) because no conclusion could be drawn from a sample of one. Even if we only tested a single movement, the time, cost, and complexity of the endeavor would render it impractical. If we were to repeat this process with multiple movements over multiple carriers and multiple routes, genuine, useful testing becomes impossible.

We will gain better knowledge and experience as these simplified standards are applied in individual cases. In that process, we can refine or improve these guidelines, as necessary, just as we have done for the Full-SAC process.[66] Accordingly, we will not delay implementation of this simplified approach.

3. Necessary Requirements for Simplification

As explained in the NPRM (at 10-11), there are two objectives of the SAC constraint: to restrain a railroad from exploiting market power over a captive shipper by charging more than it needs to earn a reasonable return on the replacement cost of the infrastructure used to serve that shipper, and to detect and eliminate the costs of inefficiencies in a carrier’s investments or operations. It is the second objective that turns the Full-SAC presentations into an intricate, expensive undertaking. Thus, the key to the ability to simplify the SAC analysis is to limit the inquiry to examine only whether the captive shipper is being forced to cross-subsidize other parts of the railroad’s rail network, and not to attempt to determine whether there are inefficiencies on the carrier’s system.

Accordingly, we must assume that the existing infrastructure along the route used for the issue traffic is needed to serve the traffic moving along that route. Moreover, we have been persuaded that the Simplified-SAC analysis should be based on the route (or predominant route) actually used for the issue traffic. Finally, the traffic group needs to include all of the traffic moving over that route. We discuss each of these requirements below.

a. Use of Existing Infrastructure

Shippers object to the inability to demonstrate inefficiencies in the carrier’s system and to the statement in the NPRM (at 11) that railroads no longer are burdened by substantial excess capacity and that the rail industry now faces the opposite situation of strained rail capacity. While they concede that the railroad industry has become more efficient in recent years, they argue that the industry is far from optimally efficient.[67]

We do not suggest that there is no excess capacity and that the railroads have the optimal infrastructure throughout the country. But we believe that the railroads, in most instances, are likely operating at a sufficiently efficient level so that it would not be worth the time and considerable expense required to attempt to measure the amount of inefficiency that could be eliminated by a SARR. As we stated in the NPRM (at 12), “even if the management of some railroads is not as economical and efficient as possible, the burden of uncovering and quantifying existing inefficiencies is so substantial as to be impracticable in all but the largest rail rate disputes.” For those cases where a shipper believes that it is bearing a significant cost due to a carrier’s inefficient configuration, it should pursue a Full-SAC case, as that is the process that can best be used to properly uncover those inefficiencies.

Moreover, in the Simplified-SAC proceeding, a complainant may submit evidence that some facilities along the carrier’s route have fallen into disuse. If compelling evidence is presented, then those facilities will be excluded from the SAC analysis. This will give the complainant an opportunity to eliminate some costs associated with configuration inefficiencies.

UP argues that railroads should have an opportunity to demonstrate that it is in the process of making new investments on the route being replicated, and to add that infrastructure into the SAC analysis. We do not agree. The purpose of the Simplified-SAC inquiry is to determine whether the captive shipper is paying more than the carrier needs to earn a reasonable return on the facilities used by the shipper. It would therefore be premature to require a shipper to provide the carrier with a return on an investment that has not yet been made and from which the shipper has not yet benefited. Once a carrier makes a new investment and builds more facilities, those facilities will automatically be captured in future rate cases. As to past cases, the carriers may seek to reopen the proceeding based on changed circumstances, but must establish the requirements for such a reopening. See Major Issues at 72-75.

Finally, we note that, even without the efficiency aspect of the SAC constraint, the Simplified-SAC approach is preferable to the Three-Benchmark approach. Simplified-SAC is more firmly grounded in CMP principles and is designed to protect shippers from abuses of market power. The Three-Benchmark approach has less of a relationship to SAC or the other CMP constraints, contains little to no means of detecting inefficiencies, and provides weaker protections against abuses of market power, particularly if the Three-Benchmark method were the only simplified rate constraint available to captive shippers.

b. Use of Existing Route

The proposal in the NPRM (at 12) would have allowed the complainant to choose the route to use for the Simplified-SAC analysis. However, some railroads argued that a complainant should be required to use the carrier’s actual route for the issue movements, or, in cases where a carrier has used multiple routes, the predominant route actually used.[68]

We are persuaded that, for the same reasons that we will require the complainant to use the actual configuration and facilities of the defendant carrier, we should also require the use of the predominant route actually used by the issue traffic during the Test Year. That is, the railroad’s route (like its configuration) is more than likely the most efficient, and even if not, any attempt to determine the most efficient route would entail too much time and expense.

There are several benefits to eliminating the issue of route selection. First, it will remove 90 days from the procedural schedule that had been devoted to route selection. Second, it will remove the cost of litigating an issue that would be unlikely to affect the outcome as we believe that in most instances carriers are using the most efficient route. Third, it significantly reduces the discovery burden on carriers, because shippers would otherwise need extensive discovery to make an informed judgment on every possible alternative routing for the SAC analysis. As with concerns over infrastructure inefficiencies, where a shipper believes that it is bearing a significant cost due to a carrier’s inefficient routing of the traffic, it should pursue a Full-SAC case, as that is the process that can best be used to properly uncover those inefficiencies.

c. Use of Existing Traffic Group

As proposed in the NPRM (at 13), the SARR’s traffic group should consist of all movements that traveled over the issue traffic route in the Test Year, and no rerouting of traffic (either issue or non-issue) will be permitted. Again, we do not believe that this will sacrifice much accuracy, as the carrier’s traffic group is likely at or near maximum efficiency.

Interested Parties cite to the ICC’s statement in Guidelines that “[t]he ability to group traffic of different shippers is essential to the theory of contestability.” Guidelines, 1 I.C.C.2d at 544. This approach will not remove their ability to group traffic; rather, it reflects grouping of traffic to the same degree enjoyed by the defendant carrier. The captive traffic will share the cost of the facilities used to serve that traffic with all other traffic that used those facilities during the Test Year. Were we to permit a complainant to base its case on the greater densities that could be achieved by routing more traffic onto those lines, or the reduced densities that would be obtained by selecting only a subset of the actual traffic, the complexity of the analysis would spiral back to that of the Full-SAC analysis. We would then need a new configuration and operating plan, and could not rely on URCS to simplify the process. Such a detailed analysis needed for rerouting of traffic or reconfiguration of the system can only take place in the Full-SAC analysis.

4. Use of URCS Data for Operating Expenses

Another necessary simplification is reliance on the defendant carrier’s URCS data, with certain modifications, to determine the operating expenses. The specifics of how URCS data will be used as a substitute are outlined in Appendix B. Both the shippers and railroads argue for changes to this proposal, and we discuss these suggestions below.

a. No Movement-Specific Adjustments

Several railroads argue for movement-specific adjustments and substitution of URCS data with actual data in several areas.[69] AAR and CP take the broad view that movement-specific adjustments should be allowed if the party wishing to make an adjustment can show that an actual cost is not captured by URCS.[70] Most railroads, however, narrow this notion to those cases where a rate challenge is brought for movements of hazardous materials (hazmat) or high-and-wide materials. They claim that we should make a special exception for cases involving those types of shipments because URCS does not accurately capture all the special costs that are associated with those movements.[71] In particular, they point to recently enacted regulations by the Department of Homeland Security, the Federal Railroad Administration, and the Pipeline and Hazardous Materials Safety Administration, that they claim will require carriers to bear additional costs for hazmat movements.[72] According to CN, some additional costs associated with hazmat movements include: the use of specialized tank cars, added insurance premiums, added safety and training costs, increased crew cost, and increased yard costs.[73]

We find that carriers’ argument flawed. URCS costs are averages of costs for a carrier’s entire traffic group. Accordingly, any higher costs associated with hazmat movements increase the average cost for the entire system. This means that to the extent the system-average costs understate the costs of hazmat movements, they overstate the costs for non-hazmat movements. Thus, when URCS is used to calculate operating expenses for a broader traffic group—and in most cases, a challenge to hazmat shipments will be merely a portion of the traffic group, because the complainant will be required to include all the defendant’s traffic that moves over the route in question—the underage for hazmat moves and overage for non-hazmat moves should to a large extent offset one another. Therefore, use of URCS system-average costs, even in cases involving a challenge to hazmat or high-and-wide shipments, should provide a reasonable approximation of the total operating expenses of the traffic group.

b. Limited Adjustments

Some carriers also argue that actual data should be used for certain “non-controversial” URCS adjustments and Phase III inputs, and they suggest that we can limit the scope of movement-specific adjustments either by providing advanced guidance as to the specific costs that can be adjusted or through technical conferences.[74] Specifically, NS/CSXT, along with CP, argue that we should allow adjustments for: fuel costs; equipment ownership; locomotive costs; crew wages; mine loading times; car hire costs; car mileage allowances; and third-party payments.[75] UP asserts that actual operating statistics should be used for: the number of locomotives; empty miles; and tare car weights.[76] In addition, UP argues that actual data should be used in place of URCS data for trackage rights payments when a SARR would use the incumbent’s trackage rights for a substantial portion of the route.[77]

The issue of whether to allow some limited adjustments to URCS was considered in Major Issues. Regarding the use of third-party payments, actual number of locomotives, and actual number of total miles or empty miles we explained (Major Issues at 58):

such piecemeal adjustments would tend to bias the results in favor of the railroads. . . . [S]elective replacement of system-average statistics – which tend to benefit the railroads – without allowing for counterbalancing adjustments that benefit shippers – which often require information not maintained in sufficient detail or at all by the railroads – may bias the entire analysis, rendering the modified URCS output unreliable.

We also addressed actual rental car payments (Major Issues at 58-59):

While we recognize this limitation in URCS, we are concerned that allowance of actual car rental costs in URCS would be subject to manipulation by the carriers. Carriers determine whether to offer an allowance at all or whether to adjust rates to reflect a shipper’s car ownership. Thus, one method of accounting for private car ownership would be deemed a “cost” in URCS while the other would not. Only railroad discretion would determine how to account for this expense.

The same reasoning applies with equal force here and, therefore, we will not allow these adjustments to URCS. As we noted in Major Issues, if a party believes that URCS could be improved, or better tailored to particular movements, it may request a separate rulemaking in which it offers its specific proposal, so that the proposal will be subjected to public comment and, if adopted, uniform application.

However, we will exclude trackage rights payments from the URCS calculation and require the defendant carrier to provide during discovery the total trackage rights payments actually made during the Test Year for all trackage rights that the SARR would use. Were we to rely on URCS, total operating costs for a route with no trackage rights would be overstated, while the costs for a route with a lot of trackage rights would be understated. Only if the SARR had the same mix of trackage rights to carrier-owned track as the defendant carrier’s entire system could we rely on URCS. Given that this is unlikely to occur, that we can easily back out the trackage rights component in URCS for all movement, and that it should be simple and inexpensive for the carrier to identify the actual total trackage rights fees over the route in question during the Test Year, we are persuaded to revise our approach to better tailor the operating expenses to the characteristics of the route and traffic examined.

c. Equipment Valuation

In the NPRM (at 51), we proposed that equipment be valued solely at book value, without adjusting for depreciation. This appeared to be the approach recommended by the Railroad Accounting Principles Board where it is impractical to develop current replacement costs. We noted, however, that SAC principles do not require that a SARR use new equipment. Rather, the SARR could use the same mix of new and used as the defendant railroad. Accordingly, we invited comment on how equipment should be valued in a Simplified-SAC proceeding. Two parties submitted substantive comments on this issue.

BNSF notes that using book value fails to account for inflation in equipment prices, thereby understating the acquisition cost of new equipment.[78] BNSF nonetheless supports the use of historical cost as a means of simplification.

AECC argues that the Board should not ignore depreciation of the railroad’s equipment when it has not provided any practical means for a SARR to account for efficiency improvements associated with the purchase of new equipment. AECC claims that, if efficiency improvements are not taken into account, the SARR should only have to pay the depreciated price of the equipment.[79]

We agree with AECC that since efficiency improvements are not taken into consideration, the SARR should only have to cover the depreciated price of using the old equipment. As explained in our NPRM, this is consistent with SAC principles, even though no complainant has provided for used equipment in any past Full-SAC case. Moreover, this is consistent with the actual equipment costs the defendant railroad bears when serving that traffic group. Accordingly, we will use book value less depreciation, as reflected in URCS, to calculate the equipment costs of the SARR in the Simplified-SAC methodology.

d. Indexing of URCS Data

BNSF points out that in some instances, there will be a time lag between the Test Year (the four most recently completed quarters preceding the filing of a complaint) and available URCS data. BNSF claims that this time lag could distort results of the Simplified-SAC analysis if, for example, traffic levels differed between the periods from which the URCS-data was derived and the four quarters of the Test Year. BNSF suggests that this discrepancy be resolved by indexing the URCS costs to the appropriate Test Year time period.

We agree that indexing of URCS costs may be necessary in some cases. The parties may introduce evidence on whether and how such indexing should take place in any individual case.

e. URCS Variability Factor

In the NPRM (at 52), we explained that, in obtaining cost data from URCS, parties should set the variability parameter to 100% for all expense categories to provide a total operating cost estimate rather than a variable cost estimate (which would exclude fixed costs that should be included in the SAC analysis). The only party to address this issue was AECC, which argues that limitations in the ability of URCS to produce relevant cost estimates should be addressed through URCS refinements, and not through ad hoc adjustments developed in rate cases.

Contrary to AECC misimpression, we did not propose any change to the manner in which URCS calculates costs. Rather, we explained to the parties the method to obtain total costs from URCS. This procedure will run URCS in the 100% Flow Thru mode for Simplified-SAC cases.

5. Use of Rolling Averages for Road Property Investment

We proposed in the NPRM to simplify the calculation of RPI costs by using the findings of the Board in prior Full-SAC proceedings. Specifically, we proposed to use a rolling average of past RPI costs, with certain adjustments for specific categories of costs, as detailed in the NPRM (at 39-48). Both the carriers and the shippers have objections to this approach as follows.

a. High-Density vs. Low-Density RPI Costs

BNSF and UP both raise the issue that taking an average of RPI costs from prior Full-SAC cases may not accurately reflect the RPI costs needed to construct a SARR in a Simplified-SAC case. BNSF notes that Full-SAC cases have involved construction of SARRs designed to handle coal traffic, whereas in a Simplified-SAC case the SARR would likely be designed to handle a different commodity or mix of commodities, and that the infrastructure cost for one may not be indicative of the cost for the other.[80] UP echoes this concern, noting that differences in geography between a coal-SARR and non-coal SARR may affect land and public improvement costs, and that a coal-only railroad would not include the facilities that would be necessary to handle a mix of coal, manifest, intermodal, and automotive traffic.[81] Accordingly, they argue that railroads should be allowed to introduce evidence on necessary adjustments.[82]

We acknowledge that there will be some cost differences between a high-density network and a low-density network, but we believe the impact of these differences would be minimal and should not affect the overall outcome of a Simplified-SAC case. We deliberately did not rely simply on the average cost per mile from prior Full-SAC cases, but instead choose to use findings on certain costs that should not change materially (like the cost of a particular type of bridge) and then permit case-specific information about those factors that would change (like the number of bridges). Moreover, some of the difference in RPI costs for a low-density network compared to a high-density network has already been accounted for in some instances by our proposed use of costs on a per-ton basis. For example, the Buildings & Facilities cost from prior Full-SAC cases is expressed on a per-ton basis.

BNSF and UP cite to several examples where the high-density RPI costs from Full-SAC cases would understate low-density RPI costs. But they ignore the fact that, for other cost items, it is likely that the Full-SAC RPI costs would overstate low-density costs. For example, on high-density networks (like the ones built for SARRs in Full-SAC cases), there is less spacing between the ties, meaning that more ties are needed than on low-density networks. Thus, using the Full-SAC RPI averages will likely overstate tie costs for a low-density network. Such overstated costs should partially offset any understated costs, further reducing any potential significance of the understated costs.

In the end, simplification comes with a cost of imprecision, and we have attempted to limit that likelihood by simplifying parts, but not all, of the RPI analysis. But if we permit the carrier to introduce evidence as to the actual RPI costs, we must permit the shippers an opportunity to do the same. This would in turn entail elaborate discovery and require hiring of more experts. And if we make the Simplified-SAC process more expensive, we would need to expand the role of the least-precise, Three-Benchmark approach. On balance, we will not adopt this recommendation at this time. Once we have observed how the Simplified-SAC method will work in particular cases, we can refine the approach to create less (or more) simplification as needed.

b. Eastern Costs vs. Western Costs

Some shippers argue that, using an average of figures from past Full-SAC cases, which includes a combination of cases involving SARRs built in the eastern portion and western portion of the U.S., would not address the fact that construction costs may vary significantly between those two regions.[83] Accordingly, some shippers suggest that the accuracy of these averages could be improved by allowing parties to divide those RPI costs into two separate groups: one for eastern cases and one for western cases.[84]

We believe the concerns about differences between eastern and western costs are, for the most part, unfounded. A review of the RPI costs in Appendix A shows that the only categories where there are differences between eastern and western cases that could have a significant effect on the results of a Simplified-SAC analysis are Roadbed Preparation and Bridges. But for Roadbed Preparation, we allow the parties to use the actual quantities of material, generally based on the ICC Engineering Reports (Engrg Rpts), to determine location-specific quantities. Thus, the potentially higher grading quantities in the mountainous parts of the East would be accounted for by using the Engrg Rpts for the specific eastern route. As for Bridges, we agree with the shippers that the differences between eastern and western costs are fairly significant, and therefore, we are modifying the procedures for calculating bridge costs to better account for the differences between eastern and western costs. The specifics of this procedure are discussed in more detail in Appendix A.

c. Verification of Data

Interested Parties object to reliance on data that the parties do not have access to. They assert that calculation of the “average absolute deviation” of the rolling RPI averages shows that the averages are not reliable indicators of actual costs. They posit that this is caused by outliers in individual case RPI totals, but cannot pursue that hypothesis.[85] They also point to our culvert expense methodology, which provides for that expense to be calculated using a linear regression model based on the type of culvert involved. BNSF counters that any outliers will either have little impact on the overall RPI costs or could be addressed by the introduction of evidence of actual costs, should we allow it.[86]

The fact that this shipper group has not seen the data from the prior Full-SAC cases does not mean that it has not been verified. In each prior case, the Board based its decision on evidence that had been tested through the adversarial process; its findings have been set forth in extensive detail in those cases; and at the conclusion of the case, the Board’s workpapers have been provided to the parties to that case for them to review and determine if errors were made; and, where a party found an error and brought it to the Board’s attention, the Board has corrected for these errors.[87]

We agree with BNSF that any outliers that may be present in the RPI averages would have a minimal impact. Interested Parties pointed to only five specific costs where the variation is significant enough to suggest outliers: residential properties, commercial properties, Type I bridges, public improvements without grade separations, and public improvements with grade separations. But of these five costs, two relate to Public Improvements, which, as BNSF notes, are the smallest category of RPI costs. And the five categories identified are just a handful in a long list of RPI costs, and should not skew the average significantly. Therefore, even if there are outliers for these cost categories, they should not significantly alter the outcome of a case.

As we noted in the September 2006 Decision in this proceeding, release of the workpapers supporting the various tables in Appendix A of the NPRM would entail the release of confidential, commercially sensitive information. Moreover, we find that it would be contrary to the goal of simplification and expeditiousness to introduce consideration of adjustments to these averages. Accordingly, we will use the proposed methodology for calculation of RPI costs in Simplified-SAC cases, as set forth in Appendix A.

6. Internal Cross-Subsidy Analysis

As stated in Otter Tail (at 24-25), a complainant cannot prove an impermissible cross-subsidy by shifting “responsibility for paying for facilities it uses to other shippers who do not benefit from those facilities.” Rather, a complainant may share the cost of the infrastructure required to serve it only with other shippers using those same facilities. Thus, to show that the captive shipper is cross-subsidizing other traffic, the evidence must at a minimum demonstrate that the revenue from the challenged rate, combined with revenue from other traffic that could share those facilities, exceeds the costs attributable to serving that traffic.

This requirement, which was first set forth in PPL and refined in Otter Tail, is known as the internal cross-subsidy or PPL test. In the NPRM (at 13), we proposed eliminating this burden for the complainant by making the test an affirmative defence, with the evidentiary burden of production and persuasion shifted to the railroad.

Several shippers argue that, even though the PPL test will now be an affirmative defense, it will still be difficult for small shippers on lighter-density lines to obtain relief. ARC/PPL notes that many small shippers do not have the volume (and thus density) necessary to survive a PPL defense.[88] In particular, ARC/PPL claims that all a railroad will have to do to win a Simplified-SAC case is focus on the portion of the issue move that has the lightest density and demonstrate that it does not pass the PPL test.[89] The difficulty in surviving a PPL defense, ARC/PPL and Interested Parties claim, would be compounded if the complainant is not allowed to reroute traffic.[90]

Despite the shippers’ objections, we affirm this aspect of the Simplified-SAC methodology. As we stated in Otter Tail (at 24), the SAC test is designed to determine if a captive shipper is cross-subsidizing parts of the defendant’s rail network from which it derives no benefit. The Simplified-SAC method is also designed to protect captive shippers from being forced to cross-subsidize other parts of the defendant’s rail network. But the captive shipper’s traffic cannot be cross-subsidizing other parts of the defendant’s network if the revenue from all of the traffic that uses the facilities is not enough to sustain those facilities. As stated by the D.C. Circuit Court of Appeals, to grant relief in such a scenario would amount to “steal[ing] from a penniless Peter to pay Paul.”[91] This reasoning is no less valid in a Simplified-SAC case than in a Full-SAC case.

7. Rate Reduction (Maximum Markup Methodology)

If the evidence reveals that the defendant carrier is forcing the captive shipper to cross-subsidize other parts of the defendant’s network, we will use the same methodology employed in Full-SAC cases to determine the maximum lawful rate. Several carriers object to the method we adopted in Major Issues for Full-SAC cases, called the Maximum Markup Methodology.[92] However, their objections are either repetitive of arguments considered, and rejected, in Major Issues,[93] or represent new legal interpretations of the statute that should have been presented to the agency in Major Issues.[94] Moreover, they have offered no persuasive reason why using the same approach in both Full-SAC and Simplified-SAC is unreasonable. We cannot responsibly continue to use the now-discredited percent reduction method in Simplified-SAC proceedings, as this approach can be easily manipulated by the railroads. See Major Issues at 14-16. Whether such manipulation would be more or less likely to occur for smaller disputes is irrelevant, as the Board must protect the integrity of its rate prescription process.

8. Computation of Rate Relief

In the NPRM (at 13-14), we proposed that a rate prescription not be calculated for all future periods in the Board’s decision. Rather, where the analysis shows that the challenged rate was unreasonable in the Test Year, the parties would be instructed to apply the same method to determine the maximum lawful rate for succeeding years, up to a maximum 5-year period. Under that proposal, the parties would be required to automatically update the analysis each year, once historical information and operating costs become available.

Shippers object to what they term an “annual true-up.” They argue that annually updating the traffic and operating cost analyses to reflect any changes would be one of the most complex and costly portions of the proposed Simplified-SAC procedures[95] and would lead to the re-litigation throughout the 5-year period.[96] The shippers propose that once a rate is prescribed, it remain in effect for the remainder of the period.[97]

Some of the railroads agree that the annual procedure could prove to be costly and time-consuming. BNSF suggests that cost-sharing discovery measures be adopted.[98] BNSF and UP also argue that railroads should be allowed to modify RPI costs in future years to account for any increases in capacity on the SARR route that have occurred since the Test Year; otherwise a shipper would benefit in the future years from the increased traffic volumes without having to help pay for the investments that made those increases possible.[99]

NS/CSXT and CP suggest that we adopt a presumption that the initial-year rate prescription should remain in effect, but allow each party to request a rate adjustment to reflect changes in traffic volume, revenues and URCS costs.[100] They argue that a party would only request an adjustment if the adjustment was worth the time and expense of conducting the process.[101]

Upon further consideration, we agree that performing a true-up each year would prove costly and for that reason could discourage a shipper from bringing a Simplified-SAC case. Moreover, we do not believe the alternative suggested by NS/CSXT and CP would avoid the undue costs. We are not as confident that the parties would generally forgo an adjustment process. A party could utilize the very expensive true-up process to impose a substantial burden on the other party for purposes of negotiating a more favorable rate. Moreover, to decide whether to pursue an adjustment, the shipper would still require extensive discovery and then perform calculations to determine the potential adjustment, which is itself a costly activity. By providing for the prescription to remain in effect for the full 5-year relief period, we believe that we can greatly simplify the process and reduce the cost of Simplified-SAC to both parties at minimal cost, as a party may file a petition for reopening if there are substantially changed circumstances. See Major Issues (at 67-75).

9. Post-Prescription Follow-ups

UP argues that a railroad should be allowed to provide evidence that the application of Simplified-SAC in a particular case has produced a result that is plainly flawed or illogical, a process that UP refers to as a “reality check.”[102] This suggestion is unnecessary. Under the statute (at 49 U.S.C. 722(c)) and the Board’s regulations (at 49 CFR 1115.3 and 1115.4), a party may file a petition for reopening or reconsideration of a decision on the grounds that we committed material error.

NS/CSXT suggest that we establish procedures for periodic, public monitoring of the effects of these new simplified guidelines, to ensure that the procedures are not generating unintended negative consequences. It is not necessary to establish specific monitoring procedures. We will remain vigilant regarding the effects of our rules on the rail industry and will revisit these rules as needed and parties may petition the Board to revise these simplified guidelines as needed.

10. Discovery

The parties raise several issues pertaining to the discovery procedures proposed in the NPRM (at 17-19) for Simplified-SAC cases. We discuss each of these issues below.

a. Disproportionate Burden

Several parties argue that the discovery burden will weigh too heavily on a defendant carrier, as it would be the party responsible for compiling and producing most of the data.[103] Several carriers also claim that the discovery burden is unfair in that it essentially requires the defendant to make a complainant’s case for it.[104] NS/CSXT specifically claim that the requirements placed on defendant carriers would be “extraordinary and undue burdens.”[105] In addition to the logistical difficulties, the railroads argue that the discovery procedures would create a policy with negative consequences. Because the discovery burden will be borne mostly by the railroads, they express concern that a shipper will be able to bring a Simplified-SAC case with very little risk and therefore would file frivolous cases as unfair leverage in rate negotiations.[106]

To address these concerns, several carriers propose that we adopt a “loser-pays” system. Under such a system, the losing party would bear the financial burden of the discovery. Thus, if a shipper loses, it would be required to pay the defendant carrier’s discovery expenses.[107] Any disputes over the amount that must be paid by the losing party could be resolved by the Board.[108] Alternatively, NS/CSXT suggest that neither party should be required to collect, develop, arrange, or otherwise generate data or information that it does not produce in the normal course of business or arrange such data other than in the manner in which it normally exists.[109]

We agree with Interested Parties that it would be difficult to monitor and apply a loser-pays system.[110] Determining how much the loser owes could become complex and could cause further litigation, contrary to the goals of this proceeding. As Interested Parties note, who wins and who loses a case would not always be black and white. Interested Parties give the example of a complainant who seeks a $2,000 per car rate reduction, but only receives a $200 per car reduction.[111] In this situation, there would be no clear-cut winner. The costs of discovery activities themselves would also be murky. As noted above, the parties significantly dispute the cost needed to bring a Simplified-SAC case, and these types of disputes would undoubtedly arise in loser-pays determinations. We can envision parties arguing that a specific discovery cost was unnecessary or cost more than it should have. In addition, under a loser-pays system, the railroads might have an incentive to inflate their costs in cases where they are confident they will prevail.

Such complications could lead to a system where the disputes over the costs of litigating a Simplified-SAC become more complicated than the case itself. Moreover, as Interested Parties note, the risk of bringing a case would be substantially increased by the possibility that the complaining shipper would have to bear unknown discovery costs.[112] This increased risk could discourage shippers from bringing cases under the Simplified-SAC model, creating a chilling effect that is directly contrary to the goal of this proceeding.

As for the carriers’ claim that placing the discovery burden mostly on their shoulders will be unduly burdensome, we note that the discovery procedures here are not significantly different from the discovery procedures in Full-SAC cases, with which the carriers have been able to comply adequately. Moreover, most of the data necessary for a SAC analysis (be it Full-SAC or Simplified-SAC) resides with the railroad, so requiring the railroad to bear more of the discovery burden is unavoidable given the nature of these cases and represents a cost of doing business in a regulated industry. At any rate, the carriers’ claim that the railroad will essentially be making a shipper’s case for it is exaggerated. A complaining shipper still must take the data that is provided by the carrier to craft its case-in-chief, as in a Full-SAC case.

Although we do not believe that there is a way that the discovery procedures could be altered so that the burden is more evenly shared, we agree with the railroads that there needs to be a safeguard that prevents shippers from bringing frivolous cases or using the threat of frivolous cases as leverage. To mitigate this concern, we will increase the filing fee for a Simplified-SAC case, as discussed below. We also believe that mediation will help to reduce the discovery burden on the railroads. Even if the case is not resolved, the mediation may resolve specific issues, and in so doing may obviate the need for the railroad to produce certain information on discovery.

b. Other Issues

BNSF suggests that railroads be allowed to file the First and Second Disclosure under protective order since the information contained in these filings could be commercially sensitive.[113] We agree that some of the information contained in these filings should remain confidential. Accordingly, we will require the railroad’s First and Second Disclosures to be filed under a protective order.

Some carriers also suggest that we should broaden the range of documents that a shipper must produce in its initial disclosure regarding market dominance. UP suggests that the shipper be required to submit all documents “related to” determination of the feasibility of transportation alternatives, rather than just those “relied upon.” According to UP, a complainant could avoid disclosing relevant documents by not “relying” on them.[114] Similarly, NS/CSXT suggest that the information that the shipper be required to produce in its initial disclosure also include all correspondence, studies, and analyses regarding transportation alternatives; a list of all shipments of the commodity at issue made by the shipper in the preceding 2 years; and any other quoted rates the shippers has obtained, including quotes for non-rail transportation.

Although in discovery matters, we are neither governed nor limited by the Federal Rules of Civil Procedure (FRCP),[115] our discovery rules follow generally those in the FRCP.[116] Under Rule 26, as part of its initial disclosure, a party must provide to the other party “a copy of . . . all documents, electronically stored information, and tangible things that are in the possession, custody, or control of the party and that the disclosing party may use to support its claims or defenses, unless solely for impeachment.”[117] The Advisory Committee Notes explain that this refers to all potentially relevant documents. But while broader initial discovery may be required in federal civil litigation, we will cabin the initial disclosures here to only information “relied upon” by the complainant. Other information within the possession, custody, or control of the complainant that is potentially relevant to the inquiry will be the subject of the discovery process.

UP also suggests that for both parties’ First Disclosure, the Board only require parties to produce workpapers sufficient to show how the party derived the input to the Phase III model, rather than “all documents that it relied upon.” UP asserts that requiring the parties to present a broader range of documents is not necessary.[118] We believe that there should be little difference between the parties’ workpapers showing how they derived the inputs and “all documents relied upon.” Without providing the underlying documentation on which the figures in the workpapers are based, the opposing party cannot verify that those numbers were correctly calculated. Without this safeguard, a party could manipulate or simply invent data that are favorable to its case. In any event, the slight additional disclosure should be neither burdensome for the party to produce nor the other party to review. Thus, preferring to err on the side of caution, we reject UP’s suggestion.

11. Filing Fee

Under the user fee schedule,[119] the filing fee for bringing a Full-SAC case is $178,200, while the fee for bringing a case under the existing simplified guidelines is only $150.[120] The filing fee for all other formal complaints is $17,600.[121] Although we update our entire fee schedule annually, we can also update any fee if the additional update is necessary.[122]

Here, we conclude that applying the $150 fee to a case handled under the Simplified-SAC approach would be inappropriate for two reasons. As noted above, this low user fee provides virtually no deterrent to a shipper bring a Simplified-SAC case against a carrier simply to engage in a fishing expedition or to use as leverage in rate negotiations. Moreover, a $150 filing fee would be vastly incongruous to the cost to the Board to process a Simplified-SAC proceeding. Under the Independent Offices Appropriation Act of 1952, agencies are directed to ensure that “each service or thing of value provided by an agency . . . to a person . . . be self-sustaining to the extent possible.”[123] Because the $150 fee currently in place for small rate complaints would be far from self-sustaining for a Simplified-SAC analysis, in addition to the dilemma it presents regarding the potential for frivolous cases, we conclude that this fee should not apply in cases brought under the Simplified-SAC methodology. However, the fee of $150 will be left intact for cases brought under the Three-Benchmark methodology.

Currently, we have a catch-all user fee of $17,600 for “all other formal complaints.” There is no reason to believe that the cost for the Board to process a Simplified-SAC case would be less than the cost to process other formal complaints. However, to ensure that the filing fee is not unduly expensive or chilling, we will reduce the user fee for processing of Simplified-SAC cases by 30%. This computation yields a user fee of $10,600.[124]

We recognize that this is significantly higher than $150 for bringing a complaint under the existing simplified procedures, but potential recovery involved in many of these complaint cases can be as much as $5 million. If the higher fee would constitute a hardship in a particular case, the complainant can request relief under the waiver procedures. See 49 CFR 1002.2(e).

12. Procedural Schedule

The procedural schedule for processing of a Simplified-SAC was set forth in detail in the NPRM (at 16-17). Shippers generally argue that an 18-month schedule is too long to be considered “expedited” and they advocate a schedule of not more than 6 months.[125] At the same time, Interested Parties argue that we cannot cut substantial time from the proposed procedural schedule without jeopardizing the ability of the parties to have a full, fair, and meaningful opportunity to present their cases.[126]

UP suggests that the procedural schedule could be shortened by having only two rounds of evidentiary filings, rather than three, with a technical conference between the first and second rounds of evidence.[127] UP notes that since the defendant in a Simplified-SAC case will not have to await the complainant’s opening evidence to know the configuration and traffic group selection (as it does in a Full-SAC case), the parties could simultaneously submit opening evidence on variable cost and stand-alone issues, participate in a technical conference, and then file simultaneous reply evidence.[128]

We conclude that reducing the rounds of evidence from three to two would not be appropriate. A three-round procedure is preferable because it provides the complainant—who bears the burden of persuasion—an opportunity to respond to issues raised by the defendant carrier. The complainant may either adjust its opening evidence for errors pointed out by the defendant on reply, accept the defendant’s reply evidence, or rebut that evidence. The third round therefore serves a useful function by providing the Board with a more complete and accurate record, as well as reducing the number of issues over which the parties disagree and the Board must resolve.

NS/CSXT argue that the 30 days afforded the railroad to gather and develop the data needed for its Second Disclosure data is inadequate and that a minimum of 90 days is needed.[129] To address this concern, NS/CSXT suggest that we shift time in the schedule devoted to the selection of the route to permit more time for a railroad to submit its Second Disclosure.[130]

As discussed above, for other reasons we have decided to require the Simplified-SAC analysis to be based on the predominant actual route used for the traffic at issue. Therefore, there is no need for 90 days devoted to the issue of route selection. Instead, the railroad will be afforded additional time to compose its Second Disclosure, and the remaining time will be devoted to the mediation process. Based on these changes, the revised procedural schedule for Simplified-SAC proceedings is set forth in the body of the decision. We will monitor how well this schedule works in practice for cases using the Simplified-SAC approach and make any adjustments to the procedural schedule as necessary.

13. Results Compared to Full-SAC

Many shippers argue that the Board should abandon the Simplified-SAC method altogether because it will unfairly penalize shippers seeking to use these simplified guidelines. They argue that any rate prescribed under the Simplified-SAC method will be higher than those that would be prescribed under a Full-SAC analysis, because of their inability to craft a SARR that eliminates the carrier’s inefficiencies. [131]

While a Simplified-SAC case may not provide the same amount of rate relief as would be achieved in a Full-SAC case (although that is not a certainty), it will not cost nearly as much to obtain some rate relief. Moreover, it would not be possible to develop a simplified procedure that fully replicates the results of a Full-SAC proceeding. If that were possible, there would be no need for anyone to use the Full-SAC method. Thus, we have tried to achieve the next best thing, which is to develop a procedure that simplifies, expedites, and reduces the cost for a shipper to bring a rate challenge, while approximating the results of Full-SAC as closely as possible. We believe that the Simplified-SAC methodology adopted here will achieve our goal.

Finally, there is no reason to believe that the results of the Three-Benchmark method, which the shippers would have us rely on instead of Simplified-SAC, would be more closely in line with a Full-SAC rate prescription. Because Simplified-SAC is more similar to the (more precise) Full-SAC analysis, the results of Simplified-SAC should be fairer, more precise, and better supported in economic principles than those produced under the Three-Benchmark approach.

II. Three-Benchmark Approach

Carriers have presented a variety of objections to the use of the Three-Benchmark approach, as proposed. Those objections, as well as modifications suggested by the shippers, are discussed below.

1. Union Pacific’s Objections to Three-Benchmark Approach

We first address UP’s opposition to any rate comparison approach. UP first argues that the expense of bringing a Simplified-SAC would be comparable to the expense of pursuing a case under the Three-Benchmark approach, and that, therefore, there is no reason to have the Three-Benchmark method. While we have lowered the cost of bringing a Simplified-SAC case as much as possible, we do not agree with UP’s assertion that it would be no less costly than bringing a case under the Three-Benchmark approach. Because our objective is to have a rate method that is affordable for even the smallest-size rate disputes, we reject UP’s argument that there is no need for the Three-Benchmark approach.

UP’s core argument relates to a railroad’s right to engage in demand-based differential pricing. It argues that railroads must be able to set individual rates at varying levels based on each movement’s unique demand characteristics, and that there is no principled basis for concluding that a rate is unreasonably high simply because it produces an R/VC ratio above the mean of comparable captive traffic. In UP’s view, rate reasonableness “must be determined by some measure other than by reference to other rates.”[132] UP argues that the proposed application of a revenue-need adjustment factor and the confidence interval around the mean do not remedy the alleged fundamental problem in using a comparison approach.[133] Finally, UP (as well as other carriers) expresses concern that repeated application of the Three-Benchmark approach would result in a ratcheting effect, whereby rates charged to captive shippers could be systematically lowered to the jurisdictional floor.[134]

In the absence of any other suitable method, we do not believe the Board is precluded from using this approach. A comparison approach can be instructive as to the reasonable level of contribution to fixed costs (the R/VC ratio) for a particular captive movement when a second, cost-based approach is also employed to constrain rail rates. We can assume that, in setting rail rates on captive traffic, a carrier will not exceed substantially the level permitted by the SAC constraint. An adjustment to R/VC levels of captive traffic is needed, however, because the rates may be priced below the SAC constraint due to market forces. We use the ratio of RSAM R/VC>180 for that purpose. Assuming that the comparison group has been drawn properly from other captive traffic with similar characteristics—and the final-tender procedures were adopted to create incentives for both parties to submit a reasonable comparison group—we believe that these adjusted R/VC ratios would fairly reflect the maximum lawful rates the carrier could charge those potentially captive movements. This is admittedly a crude adjustment. But precision must be sacrificed for simplicity, and any simplified procedures will necessarily be very rough and imprecise. Simplified Guidelines, 1 S.T.B. at 1021.

Moreover, while UP seeks to erode the protections provided to captive shippers under the Three-Benchmark approach, it offers nothing in its place. As discussed in the body of the decision, this would leave a massive regulatory gap in the protections afforded captive shippers. Most other carriers recognize that such a result would be inconsistent with good public policy or the directives and desires of Congress, and, while expressing reservations with the approach, they found the basic comparison approach acceptable.[135] If UP believes that any kind of comparison approach is unacceptable, it should provide an alternative that would provide some protections for captive shippers with smaller rail rate disputes.

We acknowledge the concern that, in theory, repeated application of the Three-Benchmark approach could have a feedback effect that could act to lower the mean for future cases. However, we do not believe that this should be a significant concern for several reasons. First, use of the mean does not suggest that every non-issue movement in the comparison group above that threshold will also be deemed unreasonably high. If those movements were the subject of a rate complaint considered under the Three-Benchmark approach, the comparison group would be different.[136] For example, a shipper challenging a 400-mile movement might look to comparable movements of between 300 and 500 miles, while a shipper with a 500-mile movement that was included in the comparison group of the first shipper’s case might itself use a different comparison group of movements between 400 and 600 miles. So while there would be some overlap, the comparison groups would likely differ unless two movements were identical in all respects—in which case, they should have the same maximum lawful rate.

Second, the potential for ratcheting will be severely constrained by the limit on the relief available under this approach. With at most only $200,000 a year in potential rate relief available to any shipper under the Three-Benchmark approach, we expect that the more restricting and binding constraint on the rates of most captive traffic will be the SAC test—in either full or simplified form. Therefore, even though this rate comparison approach would not replicate directly the results of a SAC analysis, it would import that constraint indirectly by comparing the challenged rate against rates for other potentially captive movements that are effectively constrained by some form of the SAC test.

Finally, even if every single potentially captive shipper were to seek, and obtain, the maximum relief available under the Three-Benchmark approach, this would result in a reduction in total rail revenues by less than 2.4%.[137] But for that ratcheting potential to be realized, there would have to be an avalanche of rate cases brought to the agency. If we were to witness such an occurrence, we could reassess the advisability of this approach and hold cases in abeyance in the meantime. And the revenue adequacy adjustment (RSAM R/VC>180) will have a modest, countervailing effect.

In the end, we must balance the weaknesses of this comparison approach (including the theoretical potential for ratcheting) against the intent of Congress that simplified procedures be made available to captive shippers with smaller disputes. We have taken reasonable steps to make the SAC process, in either its full or simplified form, available to captive shippers. For the remaining cases, use of this admittedly imperfect comparison approach is necessary to provide captive shippers with small disputes some practical means of obtaining rate review and relief. We cannot, as urged by UP, simply leave captive shippers with small commercial disputes to the mercy of the carriers.

Accordingly, we must offer the Three-Benchmark comparison approach for smaller rail rate disputes. We turn now to various refinements suggested by the parties.

2. Reliance on the Mean

As explained in the NPRM (at 26-28), once the comparison group is selected, the question then becomes what point in the range of those adjusted R/VC ratios represents the maximum reasonable rate for the issue movement. The NPRM stated that there were only two practical alternatives—either the mean or the highest R/VC ratio in the comparison group—as the Board could discern no reasoned basis for selecting a point in-between. The Board concluded that selecting the upper boundary of the comparison group would not be appropriate, however, as it would ignore all the other movements in the comparison group. As all movements in the comparison group should be similar to the issue movement, although not identical, the Board could find no reason to presume that the movement with the highest R/VC ratio is any more probative of the proper maximum lawful rate for the issue movement than the movement with the lowest R/VC ratio. In other words, the Board concluded that it would weight all movements in the comparison group equally, which would mean that the average was the best indicator of the reasonable rate that would apply to the issue movement. It also proposed to develop a confidence interval around the mean to address some uncertainty in the true measure of the mean R/VC level.

BNSF argues that, instead of using the mean, we should find a rate unreasonable only if it varies substantially from the norm for rates charged to other comparable traffic. Such an approach, BNSF urges, would preserve differential pricing within the comparison group. It supports its argument with the following statement in Simplified Guidelines, 1 S.T.B. 1022:

[T]he analysis that we envision would not presume that, simply because a rate produces an R/VC ratio above the average level, it is thereby unreasonable. The nature of an average number – which is produced by combining numbers that are both above and below that level – does not permit such a presumption. Rather, what we must consider is whether the resulting markup is within a reasonable range or zone.

BNSF argues that there is no reason for the Board to depart from its prior policy. Accordingly, BNSF would have us find a challenged rate unreasonable only if it is more than 1 standard deviation from the mean of the comparison group. BNSF explains that most values within the distribution (around two-thirds) will fall within 1 standard deviation of the mean.

We do not agree with the categorical conclusion reached by the Board in 1996 that one cannot draw a presumption based on the average R/VC ratio of the comparison group. If the inquiry were to determine the market value of a particular home, for example, and we had gathered a spectrum of comparable sales, we would naturally look to the average of those comparable sales as the best evidence of the market value of the home in question. Without any evidence that would suggest we weight those homes differently, the most natural, logical number to use would be the mean of the comparison groups, which would place equal weight on every observed comparable sale.[138]

While this may not be a perfect analogy, it reflects an analysis that is similar to the one involved here. Having selected comparable traffic of other potentially captive shippers, we must decide how to use those observed markups to draw an inference or presumption as to the maximum reasonable rate for the movement at issue. Each of these movements provides some indicia or evidence that should be factored into that determination. The most natural starting point is to weight all observations equally. The true maximum lawful rate might be higher, it might be lower; without a detailed cost analysis it is impossible to tell. But if every movement in the comparison group provides equally valid evidence as to the maximum reasonable rate, then a rate above the mean is more likely than not unreasonable.

By arguing that a rate should be presumed unlawful only if it is 1 standard deviation above the mean, BNSF effectively asks the Board to place more emphasis on the R/VC ratios in the upper end of the comparison group. There is no basis for doing so. The R/VC ratios in the lower end may understate the maximum lawful rate because they are more likely to reflect rates constrained by market forces rather than regulation. But rates in the upper end might overstate a reasonable rate, as those rates might themselves be unlawfully high.

Moreover, BNSF has offered no coherent explanation for why, under its theory, the Board should stop at 1 standard deviation. The presumption could be set at 2 or 3 standard deviations above the mean, or at any fraction of a deviation (e.g., 0.5 standard deviations or 1.7 standard deviations). There is no underlying rationale for its choice, other than an evident desire to have the presumption set above the adjusted mean of the comparison group.

Use of any standard deviation would create a range within the comparison group, permitting differential pricing within that comparison group. But seeking to create a range within the comparison group is improper. Because the comparison group must be drawn from potentially captive traffic, the approach already reflects a broad range of rates the carrier charges shippers; and it reflects differential pricing, as the carrier charges higher rates to captive shippers than the rates offered shippers with competitive alternatives. The objective here is not to create another range within the comparison group, but rather to use that comparison group to establish a presumption as to the maximum lawful rate the carrier can charge, when it is infeasible to conduct a cost-based analysis.

Accordingly, we will presume that a rate is unreasonable if it is above a confidence interval around the mean R/VC of the comparison group (or reasonable if it is below that level). That will be a rebuttable presumption, however, as discussed below. Parties will be permitted to present evidence to show, among other things, that the weighting should be different. Thus, a railroad may present evidence to show that the movements in the upper end of the comparison group are more indicative of the reasonable rates that would apply to the issue movement and should therefore be afforded greater weight. Similarly, a shipper could make the opposite showing. But absent any such evidence, all movements in the comparison group will be treated and weighted equally.

3. Other Relevant Factors

In Simplified Guidelines, 1 S.T.B. at 1041, the Board described how it would apply the existing Three-Benchmark procedure as follows:

[T]he three benchmarks are only a starting point for our analysis. They can and should be supplemented, as appropriate, with any particularized evidence that would qualify or modify what one or more benchmarks might otherwise indicate. We are confident that a careful analysis of these three benchmarks, together with whatever supplementary evidence is provided in a case, should enable us to meet our modest objective – to make at least a rough call as to rate reasonableness in those cases where a more precise determination is not possible.

 

In public hearings, after a decade in which these guidelines had not been used, the shipper community explained its reluctance to use those guidelines due, at least in part, to the ambiguity created by the idea that the benchmarks would constitute merely a starting point for analysis. They asked for more guidance on how this standard would be applied so that a shipper could make an informed estimate of possible outcomes before filing a complaint. Accordingly, a more formulaic approach was proposed, under which the maximum lawful rate could be readily determined once the comparison group was selected.

Both carriers and shippers now seek more flexibility in the rate standard. Railroads want a defendant carrier to be able to introduce any evidence that might bear on either the comparison group or the reasonableness of the challenged rate. For example, UP and BNSF seek the ability to present evidence of varying degrees of product and geographic competition in the selection of the comparison group.[139] UP also suggests that the result of the formula be adjusted for congested corridors.[140] CSXT and NS argue for consideration of any “relevant” evidence that will not turn the inquiry into an expensive and time-consuming battle among experts.[141] Only CP appears to favor the formulaic approach proposed.[142]

The shipper community voices concerns that cases not become overwhelmed by the introduction of endless factors, driving the cost of small disputes upwards.[143] At the same time, they want the ability to introduce evidence addressing the Long-Cannon factors and other individualized factors that would make a rate unreasonable.[144] They suggested a page limit on post-formulaic adjustments.

We will provide for more flexibility in the rate standard under the Three-Benchmark approach. As such, either party will be permitted to introduce evidence of “other relevant factors” to show that the maximum lawful rate should be higher or lower. However, the party introducing such evidence will be required to quantify this evidence, so that the Board will have an objective, transparent means of adjusting the maximum lawful rate upwards or downwards. The burden of rebutting the presumption will be on the party seeking the change. Shippers will bear the burden of demonstrating that a downward adjustment is proper, and the carrier will have the burden of showing that an upward adjustment is proper.

We share the concerns of Interested Parties that evidentiary disputes over “other relevant factors” not get out of hand and defeat the purpose of this simplified approach. Accordingly, we will set certain limits on the kinds of evidence that will be permit. Based on our experience in Full-SAC cases with product and geographic evidence and movement-specific adjustments to URCS,[145] opening the door to such evidence would unduly complicate these procedures. Accordingly, we will not permit any evidence of product and geographic competition as to specific movements or of movement-specific adjustments to URCS. We reserve the right to proscribe other categories of evidence that would lead to complex or protracted litigation or otherwise significantly increase the expense of this simplified approach.

We do not believe that a page limit is appropriate at this time. We believe the better approach is to apply strict discovery standards, under which greater emphasis will be placed on the burden imposed. Even if the information sought is relevant, we may not permit discovery if the burden is considerable. Parties are strongly encouraged to narrowly tailor any discovery request relating to “other relevant factors.” Parties seeking such evidence will have to show how the information requested is consistent with the expedited and simplified nature of this process.

4. Access to Waybill Sample

The Waybill Sample is a sample of carload waybills for shipments by all rail carriers that terminate at least 4,500 carloads or 5% of the carloads in any one state. The Waybill Sample identifies originating and terminating freight stations, the names of all railroads participating in the movement, the point of all railroad interchanges, the number of cars, the car types, the weight in tons, the commodity type, and the freight revenues. The names of the shipper and consignee are not included in the data set. Other data in the sample, however, may permit the identification of a shipper and consignee. Therefore, railroads may encrypt (mask) revenue information associated with contract shipments to safeguard the confidentiality of the contract rates, as required by 49 U.S.C. 11904.

The Board’s rules for releasing confidential information in the Waybill Sample are set forth at 49 CFR 1244.9.[146] Transportation practitioners and consultants may obtain access to the Waybill Sample to prepare testimony in formal agency or state proceedings. To obtain access, the requestor must demonstrate (1) that the Waybill Sample is the only single source of data or that obtaining the data from other sources would be burdensome or costly; (2) the data desired are relevant to issues in a pending formal proceeding; and (3) the scope of data requested is narrowly tailored to what is “absolutely essential” to the preparation of testimony.[147] In addition, strict protective rules govern the use of these data, to ensure that practitioners and consultants do not release competitively sensitive information to any railroad or shipper client.[148]

It has been the Board’s practice not to release unmasked contract revenue information to practitioners and consultants.[149] The shipper community has asked, however, that we provide equal access to this confidential information, noting that it would be unfair and create an information asymmetry if the railroads had access to the actual revenue information, while the shippers were denied access to the same.[150]

Carriers oppose the release of any unmasked Waybill data on several grounds. First, they argue that contract movements should not be included in the comparison group, which would moot this issue.[151] As discussed below, however, we do not agree that the comparison group should be limited only to traffic moving under common carrier rates. But even if we were to exclude contract rates from the comparison group, shippers will still need the unmasked revenue information to verify the Board’s RSAM and R/VC>180 calculations.[152]

Second, some railroads argue that the rate is irrelevant to the selection of the comparable traffic groups,[153] noting with approval the statement in the NPRM (at 33) that comparability should be based on “characteristics of the movement, not the level of the rate for that movement.” While this statement is true, it does not address the practical fairness concern that forcing a shipper to present its case without the same revenue information that the railroad has may bias the outcome.[154] Knowing the actual rates of movements that could go into the comparison group provides useful information, even if simply whether to settle or continue to litigate the claim. Carriers argue that the use of a “baseball” style method for selecting the comparison group offsets any advantage that a defendant carrier might otherwise have in knowing the masked Waybill Sample data.[155] While the selection process should mitigate the incentive for carriers to exploit the information asymmetry, we are not persuaded that it would be sufficient. And, in any event, we still would have the problem of calculating two benchmarks (RSAM and R/CV>180) based on confidential information to which shippers would not otherwise have access to and that has not been subject to the rigors and scrutiny of a litigation process.

Finally, carriers have failed to explain how our standard protective orders are insufficient to protect the confidentially of the Waybill Sample. Shippers have pointed out that unmasked traffic tapes have been revealed in SAC rate cases, subject to these protective orders.[156] Indeed, the information routinely produced in full-SAC cases is far more competitively sensitive than that contained in the Waybill Sample, as the actual contracts are produced, disclosing all of the terms of the agreement. UP takes the position that the standard protective orders are insufficient.[157] It argues that consultants will work closely with the complaining shipper and that the shipper may infer from the inclusion (or exclusion) of a particular movement in the comparison group that the R/VC ratio is higher (or lower) than the challenged movement. Given the variety of factors that will influence the decision of consultants to include or exclude a particular movement, we doubt that any reliable inference could be drawn by the shipper. Moreover, UP has failed to explain how knowing that one’s R/VC ratio is below or above another R/VC ratio, without knowing by how much, provides significant competitively sensitive information. The whole purpose of the Three-Benchmark approach is to determine where the challenged rate falls in comparison to other similarly situated traffic. It is thus inevitable that the shipper will discover where its rate stands in comparison with other comparable traffic. We are persuaded that protective orders will protect the confidence of the Waybill Sample.

In sum, we are persuaded that a complainant should be given access to the unmasked Waybill Sample of the defendant carrier. Accordingly, pursuant to the procedures set forth in Waybill Release, 4 I.C.C.2d at 200 and 49 CFR 1244.9(b)(4), once a shipper files a complaint under the Benchmark methodology, we will release the unmasked Waybill Sample for the defendant carrier (or carriers, in the case of a joint rate), for the 4 years that correspond with the most recently published RSAM figures. To obtain this information, however, the shipper must include with its complaint a signed confidentiality agreement. See 49 CFR 1244.9(b)(4)(v). Parties who obtain access to this information are reminded that violations of these protective orders are subject to sanctions by the Board. See 49 CFR 1244.9(g). If we conclude the violation is particularly egregious, we reserve the right to impose more severe sanctions, including disbarment for a suitable period from representing or presenting any evidence on behalf of any client in any subsequent agency matter. Improper disclosure of this confidential information is also subject to civil penalties under 49 U.S.C. 11904.

We do not find persuasive comments arguing for access to the confidential Waybill Sample data in advance of filing a complaint. See Simplified Guidelines, 1 S.T.B. at 1054-55. Therefore, we will not allow access to that data before a complaint is filed.

5. Changes to the RSAM and R/VC>180 Benchmarks

As originally formulated, these two benchmarks examine (1) how much a carrier actually earns from potentially captive traffic (R/VC>180) and (2) how much the carrier needs to charge from potentially captive traffic to earn a reasonable return on its investments (RSAM). In the NPRM (at 23-24), we proposed to change both benchmarks to reflect all traffic, rather than just potentially captive traffic. Under the proposal, R/VC>180 would be renamed to R/VCtotal, and would measure how much the carrier actually earns from all its traffic. RSAM would similarly measure what the carrier needs to charge all its traffic to earn adequate revenues.

We reasoned that, because we use the relationship between RSAM and R/VC>180 as a ratio by which to adjust the level of the comparison movements, it would be unnecessary and improper to exclude traffic priced below the 180% R/VC level. Otherwise, responsibility for the entire revenue shortfall (or the benefits from an overage) would be placed only on the traffic with an R/VC ratio above 180%. We indicated that it therefore seemed proper and more consistent with the Long-Cannon factors, at 49 U.S.C. 10701(d)(2)(A)-(C), to spread the responsibility for a carrier’s revenue shortfall among all of the traffic it carries.

Interested Parties support the proposed change on the grounds that the current calculation could show a “shortfall” from revenue adequacy even when the Board has determined the carrier to be revenue adequate.[158] This concern is misplaced, however. Notwithstanding its name (Revenue Shortfall Allocation Method), RSAM was never designed to measure the revenue “shortfall” or how to allocate such a shortfall. Rather, it was designed to measure the revenues the carrier must earn from potentially captive traffic to earn adequate revenues.

The proposed change is opposed by AAR, which argues that the existing RSAM figure is a useful indicator of how much, on average, a railroad would have to charge captive traffic to achieve revenue adequacy. AAR contends that, with the proposed change, RSAM would say nothing about what revenues the railroad needs to obtain from demand-inelastic traffic to satisfy its revenue needs.[159]

On further reflection, our proposal to shift the focus from potentially captive traffic to all traffic is erroneous. The purpose of these two benchmarks is to provide a ratio to adjust the rates in the comparison movements to reflect better the maximum lawful rates the carrier can charge captive traffic. If, for example, the railroad is not yet charging traffic enough to earn a reasonable return on its investment, this means the carrier is not engaging in the full degree of differential pricing that the law permits. The comparison rates must therefore be adjusted upwards, as they do not reflect the maximum lawful rates the carrier can charge, but rather are apparently being constrained by other market forces. Conversely, if the railroad is earning more than it needs to earn a reasonable return on its investments, the carrier is engaged in excessive differential pricing and the comparison rates need to be adjusted downwards. The question in both circumstances is how to determine how much more (or less) demand-based differential pricing is permissible.

If the Board were to change the two benchmarks as proposed, the shortfall or overage would be allocated amongst all traffic on an equal percentage basis. In other words, if there were surplus revenue, the presumption would be that all traffic is being priced too high and should see some kind of rate reduction. This kind of “percent rate reduction” approach would not be consistent with the Long-Cannon factors and the basic structure of the statute, for the reasons discussed at length in Major Issues (at 16-20). Nor would it be consistent with the methodology adopted in Major Issues for determining the maximum lawful rates in Full-SAC cases.

If there is a shortfall, this indicates that the carrier is not engaging in the full spectrum of demand-based differential pricing that the law permits because market forces prevent it from doing so. As the goal of this comparison approach is to gauge where that legal limit is (rather than the market limit), the shortfall should be allocated only to the potentially captive traffic, thereby increasing the degree of permissible demand-based differential pricing. Conversely, if the carrier is earning more than it needs to earn adequate revenues, this means it has engaged in too much differential pricing and that the rates at the upper end of the spectrum should be reduced. It does not mean that all rates should be reduced.

Accordingly, we will not adopt this aspect of the proposal. The RSAM and R/VC>180 benchmarks will continue to focus exclusively on potentially captive traffic. We will, however, revise the manner in which the RSAM and R/VC>180 figures are calculated to simplify the method and ensure that the relationship between RSAM and R/VC>180 accurately reflects whether the carrier is revenue inadequate or not. Under the revised approach, described in the body of this decision, if a carrier is revenue inadequate, the ratio of RSAM to R/VC>180 will always be greater than 1, and the opposite will always hold true for a carrier that is revenue adequate. Moreover, R/VC>180 will not be more narrowly tailored to focus on a subset of the railroad’s traffic, as that is the role of the R/VCCOMP benchmark.

6. Comparison Group

Carriers argue that contract traffic and traffic of a non-defendant railroad should be excluded from the comparison group on the grounds that such movement cannot be easily compared with a challenged common carrier movement.[160] They contend that the contract rates are a function of the terms of the contract and that information is not contained in the Waybill Sample. They also argue that the rates for traffic of another railroad should not be compared with the defendant railroad’s rates, as carriers operate different networks with different cost structures. They observe that including non-defendant traffic introduces more confidentiality concerns by requiring the Board to release confidential, unmasked Waybill information of one railroad to a competing carrier.

These carrier modifications met with mixed reviews. U.S. DOT supports the removal of contract traffic, but it opposes the exclusion of non-defendant traffic on the grounds that there might be insufficient observations in the Waybill Sample for a meaningful comparison analysis. The shipper community opposes both proposals, but expresses particular opposition to excluding contract traffic. They observe that common carrier traffic and contract traffic are increasingly similar and they dispute the notion that no contract movement can be compared with a common carrier movement.[161]

We will exclude non-defendant traffic from the comparison group because R/VC ratios of one carrier cannot fairly be compared with the R/VC ratios charged by another railroad. The reasonable level of contribution to joint and common costs (reflected by the R/VC ratio) is first and foremost a function of the amount of joint and common costs that need to be recovered. This will vary between carriers, creating inevitable and proper differences in R/VC ratios. Moreover, the reasonable degree of differential pricing one carrier can exercise is also a function of the mix of traffic; for example, a carrier with little revenue from competitive traffic will need to recover a larger share of joint and common costs from its potentially captive traffic. The wide range in needed contribution to joint and common costs is illustrated by the range of RSAM benchmarks for the Class I carriers. Accordingly, we conclude that the R/VC ratio of potentially captive traffic of one carrier provides no useful indicia of the lawful contribution to fixed and common costs for another carrier.

As for contract movements, we will not adopt the carriers’ suggestion for two independent reasons. First, we concur with commenting shippers that one cannot assume that contract rates provide no useful information as to the maximum lawful rate of the challenged movement. Second, we share DOT’s concerns that excluding contract movements from the comparison group may, in particular cases, leave insufficient movements in the Waybill Sample to perform a statistically meaningful comparison analysis. We have considered enlarging the Waybill Sample to include a larger sample of common carrier movements, but believe the considerable cost of gathering, processing, and costing a larger sample is not justified at this time. However, in judging which comparison group to select in an individual proceeding, we will consider the amount of contract traffic in the comparison group. Thus, holding everything else constant, a comparison group that consists of just common carrier traffic will be selected over a group that includes contract traffic.

Several parties ask that we address what we will do if the Waybill Sample contains no useful comparison traffic. This Three-Benchmark approach rests on the selection of a useable comparison group. If a particular movement is so unique that there are insufficient comparable movements in the Waybill Sample, we will entertain a reasonably tailored request for comparable movements from the defendant’s own traffic tapes. Such motions will be decided on a case-by-case basis, but are not encouraged, as they will expand the cost and time of pursuing relief under this simplified approach.

As explained in the NPRM (at 20), we will select the comparison group based on information contained in the Waybill Sample released to the parties at the outset of the case and other publicly available information.[162] This limitation is necessary to place the shipper on an even playing field with the carrier so that the final-offer selection process is fair. If the carrier were permitted to use information in its files, the shipper would be entitled to discovery as to all information the carrier might have that would bear on the proper comparison group. Such discovery would be very expensive. Moreover, we rely only on movements in the Waybill Sample (which generally comprises roughly 2.5% of total movements) to contain the costs of litigation. Therefore, information in the defendant railroad’s files about other comparable movements could not be used and not need to be produced. Yet the Waybill Sample does not identify the shipper of a movement, so that trying to identify the shipper involved for a particular movement in the Waybill Sample, to develop shipper-specific information from the files of the carrier, would be complicated and expensive.

While UP and BNSF maintain that carriers should be permitted to submit any evidence on the appropriate comparison group, they have offered no means of controlling the litigation costs such a proposal would inevitably create.[163] We conclude that limiting the evidence to information already in the Waybill Sample, or other publicly available information that both parties could access, is the best procedure. Parties may develop any argument based on that information.

Finally, several parties ask for more clarity on what will constitute a proper comparison group. We believe that we have provided as much clarity as we can in the abstract both here and in the body of the decision. As cases are brought, we provide more direction in those cases and will set precedent to guide future litigants.

7. No Adjustments to URCS

Notwithstanding the Board’s initial views in Simplified Guidelines that movement-specific adjustments to URCS variable costs should be allowed, upon further reflection and given our experience with Full-SAC cases, we conclude that simplified guidelines can only be achieved by adhering strictly to the URCS model to calculate variable costs. We have imposed this limitation for all rail rate cases in Major Issues (at 23-27), the reasons for which we incorporate by reference here. This policy is especially necessary and appropriate in the context of a case brought under the Three-Benchmark approach. Our experience in Full-SAC cases demonstrates how substantial the discovery and litigation over movement-specific adjustments can be, and it is imperative that we minimize costs in small rail rate disputes. Moreover, any adjustments that would be permitted would also need to be made to movements in the comparison groups, so as not to distort the comparison. But the similar movements would likely get similar adjustments, which could cancel each other out. See Burlington, 985 F.2d at 601 (“Thus, if the adjustment were made on both sides, it might well be pointless; if on only one side, it would create phony discrepancies.”).

8. Regulatory Lag – Cost and Revenue Adjustments

BNSF observes that using the Waybill Sample introduces 1 or 2 years of regulatory lag, because the Waybill Sample reflects prior market conditions.[164] As a result, the carrier claims that the proposed approach would result in setting current maximum lawful rates based upon revenue and cost data that are dated. This time-lag, it further contends, would be a particularly acute problem in periods when rate levels are increasing. UP offers similar concerns.[165]

To address this concern, BNSF recommends that we further adjust the R/VC ratios of the comparison movement by using publicly available indices. To address changes in costs, it recommends inflating the variable costs using a hybrid of RCAF-A and RCAF-U. To account for changes in rate levels, it suggests updating the revenues using a defendant railroad’s 10-Q reports (raising revenues by the average percent increase in revenues per unit reported to the Securities and Exchange Commission).

We do not believe that any adjustment to rail costs is necessary. Because the Three-Benchmark approach focuses on R/VC ratios (where price levels are reflected both in the numerator and denominator), the effects of price shifts associated with an inflationary increase in costs should be largely offset, leaving the R/VC ratios unaffected.

Nor do we believe a revenue adjustment is appropriate. Consider a hypothetical example where a carrier was revenue adequate in 2006, such that the RSAM R/VC>180 ratio shows the carrier earning 5% more from its potentially captive traffic than is needed to earn adequate revenues in that time period. We would therefore reduce the R/VC ratios of the comparison group in 2006 by 5% to more accurately reflect reasonable rates. Assume further that there is an inevitable 1-year time lag and that SEC reports show that the carrier has increased all revenues by 10% between 2006 and 2007. It does not follow that the Board should therefore increase the comparison group R/VC ratios by 10%, as those R/VC ratios already provided the carrier more than needed to achieve adequate revenues in 2006 and there is no evidence to suggest that higher rates are proper. In fact, in this hypothetical, the evidence would suggest that an opposite adjustment should be made. That is, if a revenue adequate carrier has been raising rates, then it needs less (not more) differential pricing of potentially captive traffic. When the 2007 information becomes available, the RSAM and R/VC>180 benchmarks for 2007 would change accordingly and suggest that the comparison group R/VC levels should be adjusted downward, not upward as desired by BNSF.

However, we recognize that relying on the Waybill Sample introduces some regulatory lag in the analysis. Accordingly, parties may present (as “other relevant factors”) evidence that the presumed maximum lawful rate should be higher, or lower, due to market changes not reflected in the comparison group or the average RSAM and R/VC>180 benchmarks.

9. Waybill Sample Revenue Adjustments

Interested Parties argue that the revenue field in the Waybill Sample should be adjusted to reflect rebates or other changes that are not captured in the Waybill Sample. However, they offer no feasible way to conduct such a massive undertaking within the context of a particular rail rate dispute. If they believe there are ways to improve the accuracy and use of the Waybill Sample, they are encouraged to provide their specific recommendations in a petition for a rulemaking, but broad changes to the Waybill Sample fall outside the scope of this rulemaking.

III. Eligibility

1. Summary of Public Comments

a. NPRM Proposal

In the NPRM, we sought to offer guidance as to who may expect to qualify to use a simplified approach. We noted that our directive from Congress was to “establish a simplified and expedited method for determining the reasonableness of challenged rail rates in those cases in which a full stand-alone cost presentation is too costly, given the value of the case.”[166] Thus, we sought to find a simple, objective measure of the value of a rail rate dispute. In the NPRM, the eligibility criteria were based on the Maximum Value of the Case (MVC): the maximum rate relief the shipper could obtain over 5 years if the challenged rates were reduced to the jurisdictional floor (i.e., the level at which the R/VC ratio equals 180%).

We proposed to compute the MVC by multiplying the difference between the challenged rate and the rate floor by the annual volume of the traffic at issue. The Board would then use the most recent cost-of-capital figure for the railroad industry, which the Board publishes annually, as the discount factor to obtain the net present value. If a complaint challenged multiple rates covering different origins and destinations, the Board would aggregate the MVC for each set of movements covered by the complaint. In this fashion, the MVC would equal the net present value, as of the time of the filing of the complaint, of the maximum relief that the shipper could obtain.

This proposal was designed to offer a simple, objective calculation of a case’s value based on the level of the challenged rates, the volume of traffic at issue, and the variable cost of the movements. We noted that under this proposal, without hiring industry experts, an aggrieved shipper could calculate the MVC of its case and determine which rate method would be presumed to apply. The overarching purpose was to offer clearer guidance as to who may expect to qualify to use a simplified approach.

We proposed to base a presumption of eligibility for the Three Benchmark and Simplified-SAC methodologies based on thresholds derived from estimates of what it would cost to bring a Simplified-SAC case or Full-SAC case. If the MVC exceeded $3.5 million,[167] the Board would presume that the complainant could present a Full-SAC presentation. If the MVC was between $200,000[168] and $3.5 million, the complainant could use either the Full-SAC or Simplified-SAC method, but the Board would presume it could not use the Three-Benchmark method. Finally, if the MVC was less than $200,000, the complainant could use the Three-Benchmark method. For the future, we proposed to index annually the $3.5 million and $200,000 thresholds using the Producer Price Index (PPI), which measures the average change over time in the selling prices received by domestic producers for their output. These eligibility presumptions could be rebutted based on the likely actual value of the case.

To avoid any incentive to understate the MVC, the MVC calculation would serve to limit any prospective rate relief available under either a Simplified-SAC or Three-Benchmark presentation. Rate prescriptions would be capped at the shipper’s representation of the annual tonnage at issue. For example, if a shipper represented in its complaint that the volume of traffic at issue was 100,000 tons per year, the prescription would set the maximum rate the carrier could charge for the first 100,000 tons of traffic in a particular year; the challenged rate could continue to be charged for any additional tonnage in that year. Additionally, any rate relief—whether in the form of reparations for past shipments or rate prescription for future shipments—would be limited to 5 years.

We proposed to guard against complainants disaggregating claims that could and should be brought in a single complaint to manipulate the MVC calculation. We proposed to do so by aggregating all cases brought within a 2-year period (the statute of limitations period under 49 U.S.C. 11705(c)) that involve the same defendant railroad and the same captive origin or destination. In this way, we hoped to prevent a shipper from using a more simplistic and less accurate procedure to challenge rates on movements that, if included in the initial eligibility inquiry, would not have qualified for using that procedure.

The proposed 5-year analysis period was designed to reflect the dynamic nature of the transportation markets to which a simplified approach would likely be applied. Most captive shippers operate in a fluid market environment, where their transportation needs and traffic patterns vary unpredictably. Thus, the lengthy rate prescriptions available in a Full-SAC proceeding were deemed to be of limited value to these shippers. We also noted that even large utilities have seen the value of such relief diminish, as materially changed circumstances prompt reevaluation of older rate prescriptions. Thus, we stated that rate relief that is more than 5-years distant can be too speculative to have value.

b. Public Comments on Original Proposal

Broadly, the shipper community argues that the proposed eligibility standards would deny access to the rate reasonableness process for most movements.[169] Shippers universally regard the estimates for the costs of bringing a case as unrealistically low.[170] Many argue that any eligibility standards must include a risk factor to account for the fact that a complainant might not prevail.[171] Shippers object to the aggregation rule[172] and express concerns that it would transform guidelines meant for small disputes into guidelines available only to the smallest shippers. Shippers also contend that the MVC is not relevant to whether or not a shipper will actually bring a case, but rather that the actual value of the case should guide the eligibility determination.[173] Some shippers argue that the MVC method as proposed would effectively place eligibility in the hands of the railroads as they could set rates at levels to ensure that the simplified procedures would not be available to a shipper.[174]

Federal agencies also express concern with the MVC approach. USDOT suggests that the financial amounts proposed for small and medium cases would quickly be exceeded.[175] USDA argues that to ensure that a shipper has the opportunity to present its case, the eligibility limits should be raised.[176] USDA notes that the proposed calculation of the MVC does not consider the probability of prevailing and thus overstates the probable value of a case.[177] Similarly, North Dakota contends that the MVC is not a realistic number because it is unlikely that any complaint proceeding would result in rates being reduced to the 180% R/VC level.[178] Further, North Dakota argues that, even if rates were reduced to the 180% R/VC level, the potential relief would likely be insufficient to encourage any of the eligible shippers to pursue a rate complaint, and only the largest shippers eligible for Simplified-SAC would likely pursue a complaint. North Dakota argues that the eligibility criteria should be increased substantially so that more shippers will realistically be able to bring rate cases.[179]

The railroads generally support the goal of the MVC approach as a balancing of competing goals of simplicity and accuracy.[180] They argue that the objective of this proceeding should not be an increased number of rate reasonableness complaints, but rather a process reserved for rare instances when an eligible shipper has a complaint against a market dominant carrier.[181] The railroads contend that CMP must govern the majority of rail movements and that the crude Three-Benchmark method should be used sparingly.[182] NS and CSXT argue that the actual value of a case is an unknowable figure before litigation and cannot serve as the basis for eligibility to use the simplified procedures.[183] The railroads concur with the Board’s estimate of the costs to bring a Simplified-SAC and Full-SAC case.[184] The railroads support the proposed aggregation rule.[185] UP disagrees with eligibility as a rebuttable presumption.[186]

In response to the objections raised by the shippers and several government agencies on opening, the railroads recommend that a complainant be allowed to stipulate to a higher threshold than 180 to calculate the MVC, as long as that shipper-selected higher threshold becomes the floor for any rate prescription.[187] UP argues that allowing the shipper to specify the maximum value of the case would only work if the Board continued to adhere to its aggregation rules; otherwise, shippers would be free to bring as many cases as they choose under less accurate simplified procedures and CMP would become nothing more than a historical curiosity.[188]

Interested Parties expressed interest in the railroads’ proposal, but ultimately could not support the refinement because the approach would require a complainant to make an uninformed guess as to what R/VC ratio should serve as the prescription floor.[189] Interested Parties claim that it is impossible for a complainant to make an informed decision without access to the unmasked Waybill data or discovery.

c. Refinement of the Proposal

Building on the sort of “small claims” model suggested by the railroads, in a decision served on January 22, 2007, we sought supplemental comments on a possible refinement of the initial proposal. Rather than trying to prejudge the merits of a particular case and estimate the actual value of the case, the Board suggested it might rely on the complainant to make that assessment, and simply place a limit on the total relief available under the Simplified-SAC or the Three-Benchmark approaches. To address the concern raised by shippers about a lack of information, a complainant might be permitted to amend its complaint any time up to the filing of opening evidence. If a complainant realized that more (or less) was at stake than originally anticipated, the complainant could elect to pursue relief under the more appropriate standard for the magnitude of the dispute. Under this modification, each shipper would be free to select the methodology best suited for its dispute.

In addressing this possible “small-claims” approach, the parties were invited to comment on the following related issues: whether the Board should abandon its aggregation proposal at this time, but retain discretion to address circumstances on a case-by-case basis if it found that any particular complainant was disaggregating a larger dispute into a number of small disputes in order to manipulate the agency’s processes; whether the Board had overestimated the reasonable, expected costs to litigate a Full-SAC case in light of reforms adopted in Ex Parte No. 657 (Sub-No. 1); and whether the Board had underestimated the reasonable, expected costs to litigate a Simplified-SAC case, assuming no rerouting of issue traffic. Finally, parties were asked to comment on AECC’s suggestion that the eligibility limit should provide for use of the Simplified-SAC up to the point where the value of the case is less than or equal to the expected SAC litigation costs of both parties combined. Parties were also invited to address this suggestion and whether, if the Board were to adopt a “small claims” approach, the limit on relief for disputes resolved under the Simplified-SAC should be set at twice the cost for a shipper to litigate a Full-SAC, and for disputes resolved under the Three-Benchmark approach, at twice the cost for a shipper to litigate a Simplified-SAC.

d. Supplemental Comments on “Small Claims” Refinement

This refinement was supported by Interested Parties, and provided that eligibility is based on a realistic estimate of litigation costs and risk factor.[190] Interested Parties support the idea that a complainant be able to revise its selection of relief after a complaint is filed.[191] Interested Parties also favor the indexing of the limits of relief proposed by the Board[192] and agree that a 5‑year limitation on Three-Benchmark and Simplified-SAC relief is appropriate.[193] Finally, they reiterate their position that the Board should drop the aggregation rule, wait to see if a problem develops, and then address it in a rulemaking.[194]

AECC opposes any eligibility scheme that leaves the railroad with influence over the selection of Simplified-SAC versus Full-SAC. According to AECC, this leverage enables railroads to obtain revenues above those contemplated by the statutes and by the theory of CMP, and prevents shippers from realizing the relief from Full-SAC litigation costs that motivated Simplified-SAC in the first place.[195] AECC argues that the revised proposal does not contain enough detail. But AECC supports the Board’s eligibility proposal[196] and endorses the comments of several parties advising the Board to affirmatively monitor the performance of the new methodologies.[197]

USDOT supports the refined proposal.[198] According to USDOT, the reliance on MVC to determine eligibility was inappropriate to the extent that amount could never be recovered by shippers as a practical matter. USDOT contends that allowing shippers to determine a realistic actual value of their cases would encourage more reasoned decisions about pursuing rate relief.[199] USDOT favors elimination of the aggregation rule, subject to revisiting it if there is actual evidence of manipulation by shippers, because the rule would force many shippers with multiple customer/consignee destinations to use the very expensive and time-consuming Full-SAC procedures if they contested shipments in more than one or two of their traffic lanes.

Although very similar to their own proposal, the railroads have reversed course and oppose the adoption of a small claims model.[200] They argue that it fails to advance Congress’s directive to establish a simplified method for determining the reasonableness of challenged rail rates in those cases in which a Full-SAC presentation is too costly, given the value of the case.[201] The carriers argue that CMP must continue to govern the majority of rail rates.[202] UP argues that the intent of its proposal was not to allow shippers with large cases the ability to collect smaller rate relief and a cap on future rates.[203] The railroads fear that the model invites shippers to invoke less precise rate standards in order to obtain some limited payout from rail carriers when they could pursue rate relief using Full-SAC. But some of the railroads maintain that the proposal is incomplete, noting that the proposal does not discuss how the Board would calculate rate prescriptions in cases in which a rate is found unreasonable,[204] and that more information would be needed to provide meaningful comments.[205] The railroads contend that the proposal to modify the MVC, by allowing the complainant to stipulate the rate prescription floor at some level above R/VC 180, would address the shippers concerns about MVC overstating potential relief.[206]

The railroads continue to support the aggregation rule and oppose the proposal to simply monitor potential abuse on a case-by-case basis.[207] The railroads also continue to support the Board’s estimates to bring a Full-SAC and Simplified-SAC case and the eligibility standards derived from those estimates.[208] The railroads contend that $3.5 million is a reasonable estimate of the cost to bring a Full-SAC case and that the recent changes in Major Issues will reduce the costs of Full-SAC cases.[209] And they argue that $200,000 is a reasonable estimate of the cost to bring a Simplified-SAC case.[210] The railroads oppose the inclusion of the defendant’s litigation costs as part of the eligibility consideration.[211] The railroads also dispute the need to include a risk factor in the eligibility analysis, arguing that the Board has no rational basis for assigning a particular probability of shipper success in small rate cases, and any adjustment the Board might make to account for litigation risk would be entirely arbitrary.[212]

2. Litigation Cost Estimates

Interested Parties submitted testimony on the expected litigation costs to pursue either a Simplified-SAC or Three-Benchmark case. The evidence by their expert witness represented the only litigation estimates supported by detailed, line-item descriptions. While the carriers have critiqued these estimates, they have failed to counter with any evidence of more appropriate cost estimates. Accordingly, we conclude that the estimates presented by Interested Parties represent the best evidence of record. However, we will make certain adjustments to reflect changes implemented in this decision or components we conclude are overstated or inappropriate. Based on that evidence, as adjusted, we find that it would likely cost approximately $1 million to litigate a Simplified-SAC case and $250,000 to litigate a Three-Benchmark case.

a. Simplified-SAC

Interested Parties estimated that consulting fees to litigate a Simplified-SAC case would range between $555,100 and $1,288,700.[213] They broke out this litigation estimate into 6 phases. We make the following adjustments to those estimates:

         Phase 1 (Initial Rate Evaluation): We exclude the costs associated with evaluating eligibility, reducing the estimate to between 144 and 392 hours;

         Phase 2 (Pre-Complaint Rate Evaluation): We will reduce in half the cost estimates associated with data collection, ground inspection, and analysis of existing traffic, as the issue traffic cannot be rerouted. We also exclude entirely the costs associated with identifying and analyzing potential routes (40-160 hours). This reduces the estimate to between 460 and 1,046 hours;

         Phase 3 (Eligibility): We exclude the costs associated with evaluating eligibility. And because ample time is included in Phase 1 for developing the R/VC ratios of the challenged movements for purposes of the jurisdictional showing, the time allocated to perform this analysis again in Phase 3 is duplicative. Moreover, time is allotted for a verified statement, which is not necessary for the complaint. We therefore include in this phase only the time allocated to assist the lawyers in drafting the complaint and preparing the initial disclosures. This reduces the estimate to between 64 and 104 hours;

         Phase 4 (Discovery): We exclude as unnecessary the costs associated with evidence concerning the selected route. This reduces the estimate to between 152 and 272 hours;

         Phase 5 (Evidence): We exclude as unnecessary the costs associated with evidence concerning the selected route (40-80 hours). We also conclude that the cost estimates are overstated to develop the configuration of the SARR, the traffic group and cross-over traffic, and operating expenses. The time to develop the configuration is duplicative of time already budgeted in early phases. Moreover, the estimate does not adequately account for our decision to compel the railroads to provide the traffic group and cross-over traffic in their second disclosure, as well as having costed all movements to develop the operating expenses of the SARR. Accordingly, we will cut in half the cost estimates for these three activities. This reduces the estimate to between 1,114 and 2,384 hours;

         Phase 6 (Post-STB Decision): We exclude the costs (480 to 1,240 hours) attributed to the annual true-up process. We also exclude the costs of negotiating with the railroad (40 hours) and assisting the lawyers to appeal our decision (240 hours) as improper as the focus of the statute is on the cost of presenting a case to the agency, not the cost of a potential appeal. This reduces the estimate to between 24 and 48 hours.

With these adjustments, the estimated number of hours of consulting time needed to litigate a Simplified-SAC presentation is restated to between 1,958 and 4,246 hours. Using the blended rate per hour estimate of $175 provided by Interested Parties, these hours translate into a cost range of $342,650 to $743,050, or an average of $542,850. To this estimate we add $407,138 for lawyer’s fees (75% of the consultant’s costs)[214] and $20,000 for mediation. The total estimate is $969,988, which we round up to $1 million.

b. Three-Benchmark

Interested Parties estimated that consulting costs to litigate a Simplified-SAC proceeding would range between $115,325 and $225,750.[215] They broke out this litigation estimate into 5 phases. We make the following adjustments to those estimates:

         Phase 1 (Initial Rate Evaluation): We believe that the cost estimates for this phase are too high. First, we exclude as unnecessary the costs associated with evaluating eligibility. Second, we exclude the 16-24 hour estimate to develop the unadjusted URCS costs as excessive, given that the same number of hours was already budgeted for developing the characteristics of the movement, which is the difficult part of the inquiry. Finally, we exclude the 113 to 264 hours assigned to predicting the outcome of the Three-Benchmark analysis using the Public Waybill Sample. That database provides no means of identifying the carrier nor any variable cost information needed to develop R/VC ratios, and thus would not provide a reliable source for the type of preliminary analysis anticipated by Interested Parties.[216] Excluding these costs reduces the estimate to between 36 and 56 hours;

         Phase 2 (Eligibility): We exclude the costs associated with evaluating eligibility. And because ample time is included in Phase 1 for developing the R/VC ratios of the challenged movements for purposes of the jurisdictional showing, we conclude that the time allocated to perform this analysis again in Phase 2 is duplicative. Moreover, we exclude time allotted for a verified statement, which is not necessary for the complaint. We therefore include in this phase only the time allocated to meet with the client, assist the lawyers in drafting the complaint, and preparing for the initial disclosures. This reduces the estimate to between 28 and 56 hours;

         Phase 3 (Discovery) and Phase 4 (Evidence): We make no adjustments to these phases. The cost estimate is therefore between 312 and 588 hours.

         Phase 5 (Post-STB Decision): We exclude the costs of negotiating with the railroad (8-16 hours) and assisting the lawyers to appeal the Board’s decision (10 hours) as improper, as the focus of the statute is on the cost of presenting a case to the agency, not the cost of a potential appeal. This reduces the estimate to between 18 and 26 hours.

With these adjustments, the estimated number of hours of consulting time needed to litigate a Three-Benchmark case is restated to between 394 and 726. Using the blended rate per hour estimate of $175, these hours translate into a cost range of $68,950 to $127,050, or an average of $98,000. To this estimate we add $73,500 for lawyer’s fees (75% of the consultant’s costs) and $20,000 for mediation. The total estimate is $191,500.

To this number we must add litigation expenses to reflect our decision to permit parties to submit evidence regarding “other relevant factors.” While both railroads and shippers asked for this refinement, neither addressed the added litigation expenses that might be associated with that change. Given our intent to carefully monitor that aspect to ensure that litigation costs do not get out of hand, we conclude that increasing the expected cost by roughly 30% to $250,000 provides a fair and conservative estimate of the increased cost associated with permitting evidence of “other relevant factors.” We anticipate that the actual cost to litigate a Three-Benchmark case, particularly once a body of precedent is developed to guide the analysis, should be far less than $250,000.

IV. Jurisdictional Threshold

The Board may investigate the reasonableness of a challenged rate only where the revenues the carrier receives for transporting the movements at issue exceed 180% of its variable costs of providing the service. This jurisdictional threshold for rail rate regulation also serves as the floor for regulatory relief, because the Board cannot prescribe a rate below the jurisdictional threshold.[217]

By statute, a carrier’s variable costs are to be determined using URCS with adjustment only where the Board finds it appropriate. In particular, the statute reads:

variable costs for a rail carrier shall be determined only by using such carrier’s unadjusted costs, calculated using the Uniform Rail Costing System cost finding methodology (or an alternative methodology adopted by the Board in lieu thereof) and indexed quarterly to account for current wage and price levels in the region in which the carrier operates, with adjustments specified by the Board.[218]

In other words, for this jurisdictional inquiry, Congress instructed the parties to use “unadjusted” costs calculated under URCS—the Board’s “general purpose costing system for all regulatory costing purposes”[219]—with the decision whether to permit movement-specific adjustments committed to this agency’s discretion.

The URCS model determines, for each Class I railroad,[220] what portion of each category of costs shown in that carrier’s Annual Report to the Board (STB Form R-1) represents its system-average variable unit cost for that cost category for that year. URCS consists of a series of computer programs and manual procedures organized into three phases. Phase I compiles the raw data into a useable format, and then uses statistical estimation procedures to determine the proportion of specific expense account groupings that vary with changes in the volume of activity (such as running track maintenance, which varies with gross ton-miles). These relationships are then used in Phase II to develop the unit variable costs that can be used to cost specific rail movements. Finally, Phase III permits expeditious application of these unit costs to the specific movements. This application can be performed using the Phase III program, an interactive computer program that permits the user to enter data for the specific movements under consideration.

In individual cases under Guidelines, parties have sought to make a wide variety of “movement-specific” adjustments. Shippers advocate adjustments that would have the effect of reducing the variable costs and increasing the resulting R/VC ratios, while railroads advocate adjustments that would have the opposite effect. In the past, the Board has examined each proposed adjustment to determine whether the party proposing the adjustment has shown that its proposed figure would better reflect the variable costs of serving the particular traffic at issue than the URCS system-average figure. However, as we explained in Major Issues (at 23-27), this adjustment-by-adjustment inquiry has been enormously complex and time-consuming, has required substantial discovery, and has produced a hodgepodge of results.

In contrast, calculating variable costs based solely on URCS is a quick and administratively simple process. The advance work is performed by the Board annually, and the Phase III computer program is available to the public at a minimal cost. As the central purpose here is to create expedited and simplified procedures, we proposed to use unadjusted URCS to determine the jurisdictional threshold. NPRM at 14, 28.

This proposal is opposed by several railroads. UP argued that the Board must permit some movement-specific adjustment, including payments to third parties, operating statistics, operating costs, and hazardous materials and high-wide traffic.[221] CP makes similar arguments, urging the Board to permit movement-specific adjustments relating to “unique costs that may not be captured by URCS, where such adjustments could materially affect the outcome of a particular proceeding.”[222] CP states that, at a minimum, the Board must permit movement-specific adjustments relating to fuel, equipment ownership, crew wages, switching or handling requirements, car mileage allowances, payments to other transportation providers (e.g., haulage fees or similar payments to shortlines), and costs of complying with special handling, customs, and security regulations. BNSF argues that movement-specific adjustments associated with payments to third-parties should be included, and that any movement-specific adjustments should be permitted at the end of the proceeding if the prescribed rate is at or near the jurisdictional threshold.[223] NS and CSXT also contend that the Board must permit any movement-specific adjustments if the prescribed rate is at or near the jurisdictional threshold.[224] NS and CSXT suggest that the Board publish a list of adjustments that are presumptively appropriate (a list that according to NS/CSXT must include fuel costs, equipment ownership, locomotive costs, crew wages, mine loading times, care hire costs, car mileage allowances, and payments to third parties), and permit any other adjustments, provided they are “consistent with simplified rate procedures.”[225]

Shippers, on the other hand, were supportive of the proposal.[226] They viewed this proposal as a “mixed blessing,” as there would be many cases where the shipper would benefit from movement-specific adjustments. But shippers are willing to forgo pursuit of these adjustments, if the use of URCS data is adjustment-neutral.[227] Interested Parties note that there are almost endless examples of URCS adjustments that could be proposed by the parties in any case and that, if the process begins with the railroad-preferred adjustments, it cannot end there. Interested Parties also contend that many of the adjustments proposed by the carriers are either highly contentious or outright inappropriate.[228]

We conclude that permitting the litany of adjustments urged by the carriers, whether at the outset of the case or at its conclusion, would defeat the goals of simplification that are critical to this endeavor. In Major Issues, we decided to disallow movement-specific adjustments in all rate cases.[229] We concluded that (a) the adjustments in prior cases were complex, expensive, time consuming, and required massive discovery; (b) allowing exhaustive discovery, volumes of evidence, significant consulting fees, and months of effort before parties could determine whether the Board had jurisdiction to consider the reasonableness of a rate was inconsistent with Congress’s intent to create an administratively quick and easy-to-determine jurisdictional inquiry; and (c) these adjustments were not leading to a more accurate result.

The reasons for this policy are even more compelling for cases brought under these Simplified Guidelines, given the Congressional directive that we develop a less expensive, simplified process for those cases. The evidence in Major Issues revealed that the litigation costs of such adjustments exceed $1 million. Carriers have offered no explanation for how they can support a simplified procedure for captive shippers, but insist that we permit movement-specific adjustments. Moreover, we agree with Interested Parties that we cannot permit the carriers to introduce evidence of adjustments that favor them without permitting the shippers an equal opportunity, and access to broad discovery. Shippers will be compelled to submit counter-adjustments.

NS/CSXT’s solution is to create a procedure under which either party could introduce movement-specific adjustments applicable to the issue traffic prior to a final rate prescription.[230] NS/CSXT claim that this procedure would not significantly increase the cost or complexity of these cases because: it would only be used in cases where there is a shipper win, it would be limited to the issue traffic, and the parties would only use it if it was likely to materially affect the level of the prescription.[231] BNSF offered a similar proposal.[232]

We reject the carrier’s suggestion that we simply move any consideration of movement-specific adjustments from the beginning of the case to the end. Regardless of when a movement-specific variable cost analysis is performed, it would still pose the same problems that we identified in Major Issues and would be contrary to the goal of creating a simplified process.

If the parties believe that the URCS model—which was adopted to be our uniform rail costing model for all regulatory purposes—can be improved upon or better tailored to specific movements, they should present an appropriate proposal in a request for a separate rulemaking. The proposal could then be subjected to broader public input and, if adopted, uniform application. However, there is no place for movement-specific adjustments to URCS in rate cases handled under simplified procedures.

V. Other Comments

1. Exempted Traffic

Under 49 U.S.C. 10502(a), the Board (like the ICC before it) is directed to exempt traffic from the regulatory provisions of the Interstate Commerce Act, to the maximum extent consistent with the Act. That authority has been used to exempt various broad categories of traffic from such regulation. See 49 CFR 1039. We can revoke an exemption where necessary to achieve the regulatory objectives of the statute. 49 U.S.C. 10502(d). In considering whether to revoke an exemption:

[T]he first thing we look at . . . is whether the carrier possesses substantial market power. If it does not, then there is generally no basis for revoking an exemption. If it does, then we focus on whether regulation is necessary to protect against carrier abuse of shippers as a result of such market power. Finally, in assessing whether regulation is necessary or appropriate, we address whether regulation or exemption would, on balance, better advance the objectives of the [rail transportation policy set forth in 49 U.S.C. 10101] and the interest of the shipping public overall.[233]

If the traffic at issue were part of a class of traffic that has been exempted from Board regulation, the Board proposed that the complainant would need to include with its complaint a separate request for a partial revocation of the pertinent class exemption pursuant to 49 U.S.C. 10502(d). See NPRM at 17, 30. The Board then proposed to consider the revocation request concurrent with the rate dispute. Id.

Carriers opposed this proposal, arguing that the new procedures would weaken the standards for a partial revocation and a carrier should not be burdened with with the costs of litigation until the exemption is revoked.[234]

Interested Parties, who support this proposal, argue that the railroads have misunderstood the proposal, which would not change substantive standards but only procedures.[235] Shippers also argued that considering both the rate case and the request for the partial revocation makes sense because the standard for finding market dominance (a requirement of any rate complaint) bears a close relationship to the evidence needed to support a partial revocation of an exemption to a particular movement.[236] Finally, shippers argue that the Board would retain the discretion to bifurcate the two inquiries in a particular case.[237]

While we agree with shippers that the railroads have misapprehended the proposal, we are persuaded that we should not adopt this aspect of the proposal. The class exemptions are based on prior findings that there is a sufficiently competitive market for the transportation involved that regulatory protections are not needed. The record offers no evidence that the marketplace has materially changed for any of the exempted categories of traffic since the findings were made to exempt that traffic from regulation.[238] We do not believe it is in the public interest to expose the railroads to the potentially significant burdens of rate litigation absent evidence that a partial revocation is justified. We will therefore generally hold any rate complaint in abeyance (including all rate-related discovery and the mediation process) while parties litigate the merits of the request for a partial revocation. However, we reserve the right to permit a rate case to proceed concurrently with a request for partial revocation where simultaneous review would conserve resources and expedite the matter.[239]

2. Cross-Border Traffic

CN, CP, and KCS argue that Simplified-SAC and Three-Benchmark methodologies cannot be applied to international through traffic between the United States and Canada or the United States and Mexico. CN and CP are both Canadian companies, with movements in Canada regulated by Canadian agencies. KCS, while a U.S. company, interchanges a significant amount of traffic with its sister corporation in Mexico,[240] much of which is rebilled at the Mexican border. However, KCS has begun to market and bill this interline traffic on an international through basis.

With regard to the Simplified-SAC rate methodology, the carriers argue that URCS cannot be used as a simplifying tool to establish operating and equipment costs for the Canadian or Mexican portion of the move. They argue that there are significant differences in the cost and revenue data available for domestic U.S. traffic versus that available for the portions of rail movements in Canada or Mexico. As to the Three-Benchmark methodology, the three carriers argue that it is dependent on data that is not available for cross-border through shipments because the methodology relies on URCS to develop variable costs for both the issue traffic and for movements in the comparison group. While KCS argues that the Waybill Sample does not collect data for the Mexican portion of the move, CN and CP state that the Waybill Sample data does not include a complete sample of cross-border movements, or revenue information of the Canadian portion of those movements. Therefore, CP maintains that the Waybill Sample cannot be used to identify a reliable comparison group or to generate the revenue information needed to apply the Three-Benchmark methodology in a cross-border dispute.

In light of these limitations of our URCS model, the carriers make a variety of suggestions. CN has asked the Board to wait until a cross-border complaint is filed before setting a standard for how these cases will be handled.[241] CP asks that any final regulations provide that Simplified-SAC and Three-Benchmark methodologies are not applicable to cross-border movements. CN and KCS ask that if the Board chooses to apply the simplified rate methodologies to cross-border traffic, that the Board allow movement-specific adjustments to URCS and adjustments to the Waybill Sample to account for the specific problems associated with cross-border traffic.

We conclude that our simplified proposal will apply fully to cross-border traffic.[242] For the Canadian portion of a movement, we will use the URCS data for its U.S. subsidiary to estimate the operating costs for the entire movement. Similarly, for the Mexican portion of a movement, we will use the URCS data for its U.S. counterpart, where available, or regional URCS otherwise.

Nor will we permit the carriers to exponentially increase the expense of these “simplified” proceedings by making a series of expensive adjustments to URCS. Allowing movement-specific adjustments to URCS would defeat the purpose of creating a simplified, less expensive alternative to a Full-SAC case. Although we understand these carriers’ concern that applying URCS data to the non-U.S. portion of a move may produce inaccuracies, URCS is the only costing model we have and, as the Canadian carriers state, captive cross-border traffic is minimal such that this practical approach should not have a significant effect on those carriers. We appreciate that using URCS adds imprecision to the analysis. But what the carriers ask for—complete immunity from rate challenge if the movement travels cross-border—is unacceptable. It would circumvent the intent of Congress by leaving thousands of captive shippers at the mercy of the carrier. And it would create an incentive for carriers to favor cross-border routing to circumvent regulatory oversight.

If the carriers are concerned over the inaccuracy of using URCS to cost the Canadian or Mexican portion of a movement, they can avoid this issue entirely by rebilling the traffic at the border the same way KCS does. This would permit the Board to focus just on the American portion of the movement, the only part of the movement over which we can provide rate relief.[243]

3. Class II and Class III Carriers

The Board’s proposal would not have a significant impact on small carriers that lack independent ratemaking. However, a substantial number of regional and shortline carriers have independent ratemaking authority and can set rates jointly with, or independently from their Class I partners, and handle local movements. The Board received comments from P&L, a Class II rail carrier, and ASLRRA regarding the impact that these simplified rate guidelines would have on shortlines with independent ratemaking authority.[244]

P&L and ASLRRA believe that the revised standards could adversely affect Class II and Class III carriers and create unintended prejudices against small carriers. They argue that unadjusted URCS will understate operating costs and prevent Class II and Class III carriers from recovering their fully allocated costs. URCS, according to P&L and ASLRRA, does not accurately reflect the costs of railroads that have short-haul traffic, which is typical of shortline and regional carriers. They claim that URCS allocates costs predominately on a mileage basis; whereas most shortline and regional carrier traffic is time-sensitive rather than mileage sensitive and thus not accurately accounted for in URCS. ASLRRA maintains that shorter line haul costs are underestimated because Class II and Class III railroads typically operate at speeds much slower than large railroads, thus creating higher costs per mile for many individual movements than the Class I derived unadjusted URCS. P&L argues that some of the URCS’s inputs are not carrier-specific and are derived entirely from economic data supplied by the Class I carriers, and that this further compounds the problems of using URCS for Class II and Class III carriers. ASLRRA argues that URCS also fails to reflect the allegedly higher capital costs of the Class II and Class III carriers. According to ASLRRA, most Class II and Class III railroads are privately held and do not have access to the lower-cost public equity or debt markets that are available to Class I carriers.

Accordingly, both P&L and ASLRRA urge the Board to remove Class II and Class III carriers from the scope of the proposed guidelines. ASLRRA maintains that there is no evidence that Congress was interested in Class II and Class III carriers when it mandated that the Board establish simplified procedures for cases where a full SAC presentation is too costly. If the Board includes Class II and Class III carriers in the application of these new guidelines, they ask that the Board permit movement-specific adjustments to reflect the actual costs of service.[245]

We find all of these suggestions inappropriate where the movement involves the participation of a Class I carrier as well. Shortline and regional railroads originate or terminate roughly one quarter of all domestic rail traffic. Exempting that traffic from these guidelines would have a profound impact. Moreover, it would create perverse incentives for carriers to include a shortline in the routing to avoid these guidelines altogether. Where the shortline railroad’s participation represents only a part of a larger movement over a Class I carrier, the imprecision noted by the shortlines should have a minimal impact on the rate analysis. Because experience has shown that an attempt to develop more precise URCS data for a shortline is too burdensome and protracted, use of regional URCS data is an acceptable compromise.[246]

We will not, however, apply the Three-Benchmark approach to a purely local movement of a Class II or Class III carrier. Two of the three benchmarks needed for that approach would not be available without analyzing the traffic tapes of the shortline (to calculate the R/VC>180 benchmark) and performing a revenue adequacy inquiry (to derive RSAM benchmark). We see no way to modify the Three-Benchmark method to render it suitable for purely local shortline movements without increasing the cost to near that of a Simplified-SAC presentation. For a purely local movement of a Class II or Class III carrier, the complainant may use the Simplified-SAC approach.

We recognize that there can be a spectrum of movements between the two extremes, where the movement is dominated by a Class I carrier and where the movement is local to the Class II or Class III carrier. If the shortline or regional railroad does not want its traffic subject to the Three-Benchmark approach where the Class I’s participation in the movement is minimal, it can establish a separate challengeable local rate to the interchange point with the Class I carrier.

4. The Kansas City Southern Railway Company

KCS has also asked that these simplified approaches not be applied to movements on its system. KCS, which is a Class I carrier, asserts that it functions more like a regional rail carrier. It asserts that it does not use URCS for any costing purposes and that the proposed rule wrongly presumes that all carriers are sufficiently acquainted with URCS to be able to respond quickly to a rate case. KCS argues that, while inputs from its R-1 report into URCS may be accurate, the way URCS uses those inputs and allocates costs among various costing categories produces inaccurate results, especially for carriers with a lot of time-intensive activities. Thus, it argues, using unadjusted URCS disadvantages smaller rail systems, like itself, that have shorter lengths of haul and a higher proportion of expenses derived from switching, pickup, and delivery of freight. KCS also maintains URCS underestimates the cost of capital for KCS by using an industry-wide average cost of capital based on the four large U.S.-based Class I carriers. KCS states that its weighted average cost of capital is estimated to be in the 14-16% ranges, in contrast to the lower industry average of 12.2%.

We will not exempt KCS from these procedures or treat it as anything other than a Class I carrier. The Board maintains the data to generate system-average URCS costs specific to KCS, based on audited filings by KCS, and those costs can be easily and readily obtained. Therefore, to the extent that KCS’s traffic has shorter lengths of haul and higher switching costs, these higher costs are incorporated into URCS for KCS. URCS calculates the system-average cost of activities by dividing the cost of those activities, as reported in the R-1 report, by the volume of those activities, which are also reported in the R-1 report. There is no understatement or overstatement. Neither the cost of activities nor the volume of activities of other Class I railroads affects the costs calculated by URCS for KCS. If KCS believes that the industry-average cost of capital used by this agency should be replaced with a carrier-specific cost of capital, it should advise the Board on how to address that issue in the ongoing inquiry into how to calculate the cost of capital f