When a shipper files a formal complaint that a railroad's rate is too high, the Surface Transportation Board (Board) must determine whether the challenged rate is reasonable. To present its case using the Board's procedures for small cases, the complaining shipper needs to obtain from the Board confidential information that the Board collects regarding the rates that the defendant railroad charges other shippers for similar shipments. Pursuant to the notice of proposed rulemaking published in the Federal Register on October 27, 2010, the Board is formalizing its rules with respect to the Three-Benchmark methodology for adjudicating simplified rate case complaints, making the most recent four years of this confidential information available to parties and permitting the parties to use any combination of the four years of confidential information when presenting their cases.
Effective March 12, 2012.
FOR FURTHER INFORMATION CONTACT:
Scott Zimmerman at (202) 245-0386. Assistance for the hearing impaired is available through the Federal Information Relay Service (FIRS) at 1-800-877-8339.
The Board is formalizing its rules with respect to the Three-Benchmark methodology used to adjudicate simplified rate case complaints. Under the rule we are adopting here, the Board will release to the parties in Three-Benchmark proceedings the unmasked Carload Waybill Sample data (Waybill Sample data) 
of the defendant carrier for the four years that correspond with the most recently published Revenue Shortfall Allocation Method (RSAM) figures. The parties may then form their traffic comparison groups by choosing the movements from the released four-year Waybill Sample data that they believe are the most comparable to the issue movements.
In Simplified Standards for Rail Rate Cases (Simplified Standards), EP 646 (Sub-No. 1) (STB served Sept. 5, 2007), aff'd sub nom. CSX Transp., Inc. v. STB (CSXT I), 568 F.3d 236 (DC Cir. 2009), vacated in part on reh'g, CSX Transp., Inc. v. STB (CSXT II), 584 F.3d 1076 (DC Cir. 2009),the Board modified 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 Three-Benchmark method, originally promulgated in 1996,
compares a challenged rate of the “issue traffic” (the traffic at issue in the case), measured as the ratio of the traffic's revenues to variable costs (R/VC ratio), to the R/VC ratios of a comparison group of traffic (R/VCCOMP) drawn from the Waybill Sample data of the defendant carrier.
Under the Three-Benchmark method as revised in Simplified Standards, each party creates and proffers to the Board a proposed comparison group (R/VCCOMP), and the Board selects the one that it concludes is most similar in the aggregate to the issue movements. The Board then applies a “revenue adequacy adjustment” (the ratio of RSAM ÷ R/VC>180) to each movement in the comparison group and calculates the mean and standard deviation of the resulting R/VC ratios. If the challenged rate exceeds a reasonable confidence interval around the estimated mean, it will be presumed unreasonable, and, absent any “other relevant factors,” the maximum lawful rate will be prescribed at that boundary level.
The rule proposed in Simplified Standards would have required parties to draw their traffic comparison groups from the most recently available one year of Waybill Sample data derived from the defendant carrier's shipments of non-issue traffic. Simplified Standards, slip op. at 32-33 (STB served July 28, 2006) (Notice of Proposed Rulemaking). The final rule, however, allowed parties to form comparison groups using Waybill Sample data from the four years that correspond with the most recently published RSAM figures. Simplified Standards, slip op. at 80.
On judicial review, the court concluded that the Board had failed to provide adequate notice of the final rule regarding the available date range of Waybill Sample data. Accordingly, the court vacated that portion of Simplified Standards. CSX II, 584 F.3d at 1078. As a result, there is currently a gap in the Board's rules; i.e., there is no defined period for which unmasked Waybill Sample data is to be released in a Three-Benchmark proceeding.
On April 2, 2010, the Board issued a notice of proposed rulemaking for a rule that would provide to the parties in Three-Benchmark proceedings the unmasked Waybill Sample data of the defendant carrier for the four years that correspond with the most recently published RSAM figures. The parties would then draw their comparison groups in any combination they choose from the released Waybill Sample data. The Board received comments on this proposal from shippers, rail carriers, the U.S. Department of Agriculture, and other interested organizations.
AAR, CP and NSR/CSXT expressed concern that the Board did not provide the rationales and regulatory objectives behind the proposed rules. In response, on October 22, 2010, the Board published a revised notice, which proposed rules identical to those proposed on April 2, 2010, and included an expanded explanation of the rationales and regulatory objectives behind the proposed rules. Following publication, the Board received additional comments from rail carriers, shippers, and other interested organizations.
Although the final rules adopted in this decision are identical to those published in the two previous notices, the Board responds in further detail to the comments received in response to the April 2, 2010 and October 22, 2010 notices.
AAR and the commenting rail carriers object to permitting shippers to draw their comparison group from the four most recently available years of Waybill Sample data, because of what they characterize as “regulatory lag.” 
They argue that even the most recent one year of Waybill Sample data is unlikely to reflect current market conditions because the data may be up to two years old by the time the Board publishes the Waybill Sample. They contend that the proposed rule increases the likelihood of distorted comparison groups and results by permitting parties to use six-year old data.
AAR further contends that the Board can address any issues of data insufficiency in individual cases from the one-year data release by requiring the carrier to provide its traffic tapes for all movements of the commodity at issue for the current period.
Shippers, on the other hand, generally support adoption of the four-year Waybill Sample data rule. They argue that using multiple years of Waybill Sample data will smooth out the effects of short term variations in prices and costs that make up the data. They also claim that it is necessary to permit the use of four years of Waybill Sample data because a single year's traffic may not contain sufficient data from which to derive meaningful or representative comparison groups. Shippers maintain that the Board should require, rather than merely permit, parties to incorporate data from each year of the current four-year Waybill Sample data in developing their R/VCCOMP comparison groups, because the two other benchmarks (RSAM and R/VC>180) are calculated using Waybill Sample data for the same four-year period.
Discussion and Conclusions
Parties in a Three-Benchmark rate case may submit a comparison group from the four-year Waybill Sample data we provide them at the beginning of the case. This rule simply defines the range of data that will be available to the parties; it does not dictate how the data will be used. We are not imposing a rule that forces the parties to submit a comparison group that includes movements from each year of the four-year period, or just from the first year, or the last year, or any particular combination of years. Parties may construct their comparison groups from any combination of movements drawn from the four-year Waybill Sample data. We will continue to use the final offer selection process to select the best comparison group on a case-by-case basis.
We have three reasons for adopting this rule. First, this rule provides the parties the flexibility needed to tailor their comparison groups as they see fit. In some cases, a shipper might believe it needs to use more than one year of data to demonstrate that rates for the issue traffic were unreasonably high. Thus, a party may, for example, select its comparison group from data across all four years and argue that a group selected from all four years is the most comparable to the movements at issue. On the other hand, a party may select its comparison group from a single year's data and argue, based on that case's facts, that the best comparison group is one drawn from only that year. The Board remains the ultimate arbiter in each case of which litigant's comparison group it will use to judge the challenged rate.
Second, permitting the parties to draw a comparison group from the four-year Waybill Sample data should provide enough observations to draw a valid inference about the maximum lawful rate. One year of data may in some cases be insufficient to provide a meaningful benchmark for comparison purposes. The Board was particularly concerned in Simplified Standards with having sufficient movements of certain hazardous cargoes (known as toxic inhalation hazards or “TIH”) for parties to develop appropriate comparison groups, but our concern about data sufficiency is broader than that. As USDA noted in its comments (at 3), for example, because production of some specialty crops may vary significantly from year to year, shippers of such crops must have the flexibility to draw upon data generated during multiple year periods.
The rail carriers argue that, instead of permitting the use of four years of Waybill Sample data, we should instead require the carrier to make available its most recent traffic data. Using the most recent traffic data would, according to the carriers, meet the Board's desire for both flexibility in the selection of the comparison group and enough observations to make an informed decision.
We disagree. Based on our experience in Stand-Alone Cost (SAC) cases and in processing the annual Waybill Sample data, we have already concluded that using the prepared Waybill Sample data is one of the linchpins to the simplified rate review process. The release of four years of Waybill Sample data to the parties minimizes the possibility that additional traffic data will be needed for the parties to develop their comparison groups.
Moreover, the costs and delays associated with the collection, preparation, production, verification, and use of the carrier's most recent traffic data run contrary to Congress's directive and the Board's objective of devising simplified procedures for use in small rate cases. Because relief in Three-Benchmark cases is limited, the costs associated with extensive discovery could significantly offset, or even eliminate, any rate reduction benefits from such cases and deter shippers from seeking relief. For example, relying only on data provided by the carrier presents the problem that, unlike the Waybill Sample data, the traffic data provided by the carriers would not include the variable cost data necessary to determine R/VC ratios.
Adopting the carriers' proposal would substantially increase the cost of bringing a Three Benchmark case and impede shippers' ability to seek relief for smaller disputes.
Third, making four years of data available is fully consistent with the basic idea behind the Three-Benchmark approach. As the Board stated in Simplified Standards (at 73), in the absence of any other suitable method, 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. The Three-Benchmark methodology embodies this approach: it is a comparison-based methodology that applies a cost-based adjustment— the ratio of RSAM ÷ R/VC>180 —to the comparison groups. The Three-Benchmark method begins with the assumption that, in setting rail rates for captive traffic, “the carrier will not exceed substantially the level permitted by the SAC constraint.” Id. An adjustment to the R/VC levels of captive traffic is needed, however, because the rates may be priced below the SAC constraint due to market forces. Id. Applying the RSAM ÷ R/VC>180 adjustment factor to the R/VC ratios of the comparison group adjusts those ratios to those that would be needed for the carrier to achieve revenue adequacy.
Assuming that the comparison group has been drawn properly from other captive traffic with similar characteristics—and the final offer procedures were adopted to create incentives for both parties to submit a reasonable comparison group—we concluded that “these adjusted R/VC ratios would fairly reflect the maximum lawful rates the carrier could charge those potentially captive movements.” Id. Accordingly, 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.” Id. at 18.
The rail carriers argue against using four years of Waybill Sample data because, they claim, (1) The data will be too stale, (2) the R/VCCOMP benchmark should have no relationship to the time period used to calculate the other two benchmarks, and (3) in calculating the R/VCCOMP benchmark, there is no need to smooth out business variations in the pricing of similar traffic. The carriers also claim the proposal is flawed because rates and costs in the industry and for specific commodities change over time. These objections are best summarized by NSR and CSXT, both of which declare that “the goal of the R/VCCOMP is not to smooth out annual variations; it is to reflect as accurately as possible current market conditions in which the carrier establishes the challenged rate.” NSR/CSXT Supp. at 6-7.
The carriers' arguments are not persuasive. The fundamental purpose of the Three-Benchmark approach is not to reflect a snapshot of current market conditions; it is to use the three benchmarks to decide the reasonable maximum contribution to joint and common costs for the issue movement where no cost-based approach is feasible. The R/VCCOMP benchmark is used to approximate the maximum reasonable rate that a rail carrier could charge under the SAC constraint. The Three-Benchmark method compares the R/VC ratios (i.e., percentage markups over variable cost) of particular current movements against the R/VC ratios of comparable movements selected from any mix of movements within the four years of Waybill Sample data.
One weakness in employing this benchmark to protect shippers from unreasonable rates is that the constraint may not always approximate the maximum reasonable rate under the SAC constraint, particularly over relatively short observational periods.
By giving parties the opportunity to select their comparison groups from as much or as little data as they choose from within multiple years of Waybill Sample data, the Board can have greater confidence that the adjusted R/VC ratios of the comparison group (R/VCCOMP) selected through the final offer process will approximate the maximum reasonable level permitted by the more precise SAC constraint.
Moreover, we use the parties' comparison group to prescribe the maximum lawful rate not just at the moment a carrier's rates are challenged, but for a five year period. The maximum lawful rate for a movement (i.e., the maximum reasonable contribution to joint and common costs expressed as an R/VC ratio) may change from year to year, as it is a function of the amount of joint and common costs that need to be recovered, as well as the level and the mix of traffic, and the revenue generated by that traffic. See Simplified Standards at 82. For example, a carrier with little revenue from competitive traffic in a given year will need to recover a larger share of joint and common costs from its potentially captive traffic, id., while in a boom year when the carrier enjoys stronger revenues from competitive traffic, a carrier would need to recover less from its potentially captive traffic. It is therefore reasonable to permit parties broad latitude to draw information about the R/VC levels charged to comparable traffic from any or all of the most recent four years of Waybill Sample data for all three benchmarks. Again, the parties may argue that the circumstances of a particular case caution against drawing information from a four-year time period, or that a comparison group drawn from, say, only one or two years of Waybill Sample data is superior to one drawn from four years of data because of other characteristics of the selected movements,
or that, due to the inevitable regulatory lag, a further adjustment to all three benchmarks is needed (so-called “other relevant factors”).
We reiterate that the Board remains the ultimate arbiter of which litigant's comparison group it will use to assess the challenged rate(s), and the Board will consider the extent to which a party's comparison group is most similar in the aggregate to the issue traffic on a case-by-case basis. The final offer process gives both parties the opportunity to convince the Board that its comparison group is most similar to the issue traffic.
In addition, complainants should have access to multiple years of data so that they can make year-to-year comparisons of rate changes to identify potentially unreasonable carrier pricing behavior. Although the R/VC ratios of the issue traffic might well be similar to the R/VC ratios of comparable movements in the current year, they might be dramatically higher than the R/VC ratios of comparable shipments from prior years. We see no reason why a complainant should be deprived at the outset of the case of readily available Waybill Sample data needed to make that case.
Finally, NSR and CSXT argue that 49 U.S.C. 10701(d)(1) compels us to use the most current data when evaluating the reasonableness of rates. They maintain that the statute “requires at a minimum that the comparison group movements reflect the same market conditions that exist when the railroad established the challenged rate.” NSR/CSXT Supp. at 7. Put differently, they argue that when asked to judge the reasonableness of a rate set in 2010, we cannot perform an analysis of whether the rate was comparable to rates from 2005-2008. Id.
This statutory argument is unpersuasive for a number of reasons. First, the statute contains no such directive. Second, when judging the reasonableness of a particular rate, we routinely look to information beyond the year when the rate was established. For example, our SAC test does not judge the reasonableness of the challenged rate by looking only at a snapshot of the current financial circumstances. Rather, the SAC test requires a 10-year analysis that is structured to reflect the variations in the business cycle. See Major Issues In Rail Rate Cases, EP 657 (Sub-No. 1), slip op. at 61 (STB served Oct. 30, 1996). Some of the variables it takes into account are the annual tonnage fluctuation, change in tax laws, equity investor expectations, and inflation in the prices of the assets utilized by the industry. Coal Trading Corp. v. B&O R.R., 6 I.C.C.2d 361, 411 (1990). Third, in their example above, the Three-Benchmark approach would not compare the rate set in 2010 against the rates from 2005-2008; it would judge the reasonableness of the challenged rate by comparing the R/VC ratio (the level of contribution to joint and common cost) against the adjusted R/VC ratios of comparable traffic from 2005-2008. Finally, in a rate case, we are not asked to determine the maximum lawful rate on the day the tariff was issued, but for a multi-year prescriptive period.
This decision will not significantly affect either the quality of the human environment or the conservation of energy resources.
It is ordered:
1. The Board will adopt the rule as set forth in this decision.
2. This decision is effective on the day of service.
3. This decision will be published in the Federal Register.
Decided: March 8, 2012.
By the Board, Chairman Elliott, Vice Chairman Mulvey, and Commissioner Begeman.
[FR Doc. 2012-6551 Filed 3-16-12; 8:45 am]
BILLING CODE 4915-01-P