Modernizing Freight Rail Regulation (2015)

5

SUMMARY ASSESSMENT AND RECOMMENDATIONS

This chapter provides a synopsis of the report’s background and context discussion, an overview of recent trends in freight rail rates and issues pertaining to rail service quality and capacity, and critiques of several major elements of the federal railroad regulatory program. Together, they respond to the request by Congress for an independent examination of the

• Performance of the nation’s major railroads with regard to service levels, service quality, and rates;
• Projected demand for freight transportation over the next two decades and the constraints limiting the railroads’ ability to meet that demand; and
• Effectiveness of public policy in balancing the need for railroads to earn adequate returns with those of shippers for reasonable rates and adequate service.

The committee was asked to make recommendations on “the future role of the Surface Transportation Board [STB] in regulating railroad rates, service levels, and the railroads’ common carrier obligations, particularly as railroads may become revenue adequate.” Therefore, at the conclusion of the chapter a number of actions to end or modify long-standing regulatory requirements and practices are recommended. In many cases the actions are accompanied by proposals to replace the requirements with alternative approaches consistent with the policy goals of the Staggers Rail Act.

The theme that unifies the recommendations is modernization. The Staggers Rail Act was successful in enabling the development of an efficient, innovative, and financially strong freight railroad industry,

to the benefit of shippers and consumers. The law allowed the industry to modernize. However, many elements of the 35-year-old law’s regulatory program are not appropriate for today’s railroad industry. It is time to bring the regulatory program into the modern age.

STUDY CONTEXT

Regulatory Reforms of the Staggers Rail Act

The Staggers Rail Act of 1980 made fundamental changes in the federal railroad regulatory program. When the act was passed, the country’s private freight railroad industry was in financial and physical decline. It had been overregulated and had lost large amounts of traffic to trucks. Some railroads were receiving government subsidies, and the industry’s nationalization was a possibility. Just a few years earlier the federal government had nationalized passenger rail service. The railroads, which had once dominated the transportation of freight and passengers, were left with an asset base that had become oversized and misaligned with demand. They were generating too little revenue to pay for basic upkeep, much less reinvestment. Reducing expansive networks and other legacy capacity that had become uneconomic and making more intensive use of the capacity that remained were critical to the industry’s survival as a private enterprise. The Staggers Rail Act sought to enable such changes, which had been hindered for decades by the federal regulatory regime.

The Staggers Rail Act ended restrictions on rate setting that had made price competition among railroads nearly nonexistent and that had detracted from the ability of railroads to compete with barges and trucks. The act allowed railroads to shed excess capacity and to concentrate traffic on fewer lines using fewer locomotives and workers. Regulators were required to be more accommodating of railroad requests to discontinue unprofitable service and to sell or abandon lightly used lines. Provisions in the law were intended to hasten what had become a slow and uncertain merger review process. Regulators were instructed to conduct the reviews according to a “public interest” standard that balanced the competitive effects of the merger against other potential outcomes. One outcome to be appraised was

whether the proposed merger would enlarge the traffic base and earnings potential of the merged railroad and thus make it more financially viable (and thus less likely to need public subsidies).

The law allowed railroads to cancel many legacy trackage rights, terminal access, and reciprocal switching agreements—some imposed years before as merger conditions—that had hindered their ability in a deregulated setting to provide efficient direct service and to obtain pricing leverage over traffic that originated and terminated exclusively on their networks. Railroads would thus have an enhanced ability to concentrate traffic in more efficient movements and to charge “captive” shippers rates corresponding to their willingness to pay. Although the law permitted regulators to order access agreements if they were deemed “necessary to provide competitive rail service,” such interventions were not required, and regulators ordered them only on rare occasions out of respect for the law’s interest in revenue adequacy.

Congress’s new regulatory policy was to allow “competition and the demand for services to establish reasonable rates for transportation by rail.” In addition to ending most regulatory mechanisms that restricted pricing, the Staggers Rail Act legalized confidential contracting. With more operating and pricing freedom, railroads had more opportunity to innovate and compete for traffic that had been shifting to trucks. The new freedoms fit in with the act’s provisions that made it easier for railroads to cancel interchange agreements and retain control over sole-served traffic. A railroad could thus obtain more market power over captive shippers and exercise that power more effectively by charging them up to their maximum willingness to pay. To retain the business of other shippers having more options, the railroad was free to negotiate individual contract rates commensurate with each shipper’s willingness to pay. With private contracting made legal, the tariff rate no longer applied to all traffic. The railroads could maximize profits and prioritize profitable traffic flows without leaving any profit-generating traffic unserved.

The Staggers Rail Act reforms were thus oriented toward inducing efficiency, providing incentives for innovation and capital formation, and creating a renewed railroad alternative to trucks. The revenue-enhancing aims of the reforms were also clear. The private railroads were earning too little to keep producing and reinvesting in

their capital-intensive systems, and the law was designed to alleviate that situation. The prolonged failure of railroads to cover their cost of capital promised service losses to shippers generally, with the greatest harm to those having the most dependence on rail. The general loss of freight rail service was not a policy option, however. Billions of dollars had been spent by the federal government on subsidies to failing private railroads, with the prospect of billions more. That course was not considered desirable, so Congress structured the Staggers Rail Act to emphasize railroad revenue adequacy.

The Staggers Rail Act was explicit that railroads should be allowed to earn their cost of capital and required regulators to make annual assessments of each railroad’s progress in doing so. The assessments were to aid policy makers in monitoring the industry’s financial condition and in determining whether additional actions were needed to turn the industry around. Revitalization was the immediate goal, followed by ensuring the long-term viability of a privately owned and operated freight railroad system. Whether the required annual revenue adequacy assessments were meant for other purposes, perhaps even to monitor the deregulated industry for signs of excessive profits, is unclear. Any purpose other than monitoring the rescue effort appears to have been secondary when the Staggers Rail Act was passed.

Unlike the comprehensive deregulation laws that were enacted contemporaneously for other transportation industries, the Staggers Rail Act preserved a number of economic regulations and added requirements intended to protect the interests of shippers who might be harmed by the regulatory reforms. A tenet of railroad regulation arising from common law is that railroads are “common carriers,” obligated to provide service to all shippers on reasonable request and “without discrimination.” This obligation was preserved for all traffic that could not be moved competitively by truck. In addition, Congress appears to have recognized that features of the law would enable railroads to obtain more market power and exploit it by raising rates, particularly for shippers who would lose transportation options when railroads elected to cancel their legacy interchange agreements with competitors. The law therefore did not end rate regulation entirely but included provisions allowing shippers using common carriage in markets lacking effective competition to challenge a rate as being unreasonably high.

While the Staggers Rail Act preserved the duty to provide common carrier service, it did not define either the minimum level of service that should be supplied or the maximum rates that could be charged. Minimum service requirements had never been defined for common carriage by rail, whereas rates had long been subject to regulation. With rates deregulated and allowed to change, service terms and levels could be expected to change as well. The Staggers Rail Act simply required that all common carrier rates be “reasonable.” The reasonableness standard in railroad regulation had evolved from common law notions of fairness implying that similarly situated shippers should pay generally similar rates. The Staggers Rail Act did not prescribe a new or specific fairness standard. It declared that all common carrier rates should be presumed reasonable and that only shippers in markets in which the railroad qualifies as being “dominant” could challenge a tariff rate on grounds that it is unreasonable. This exception to the law’s pricing freedoms could be viewed as a counterbalance to the law’s allowing railroads to obtain and exercise more market power as a means of covering capital costs, regaining financial viability, and ending the demand for public subsidies.

Aftermath of the Staggers Rail Act Reforms

Many of the reforms in the Staggers Rail Act could be adopted fairly quickly. Confidential contracting became legal immediately, and large amounts of traffic in markets in which it became heavily used were effectively deregulated. The Interstate Commerce Commission (ICC), the industry’s regulatory body, exempted entire categories of truck-competitive commodities and car types from regulation, including shipments moved in boxcars and intermodal containers. The agency expedited railroad requests to sell and abandon lightly used lines and approved applications for mergers involving railroads having overlapping networks on the basis of the public interest standard.

The reforms had an immediate impact on the subsidy debate: no federal operating subsidies were granted after 1981. Other changes occurred quickly as well. By 1995, the major railroads had shed more than 40 percent of their track mileage, two-thirds of their employees, and about one-third of their locomotives compared with 1970, when

the wave of railroad bankruptcies had commenced. During the period, railroads had learned to make much more intensive use of their remaining capacity; ton-miles per track mile tripled, ton-miles per carload nearly doubled, and tons per train grew by nearly 60 percent. Hundreds of new regional and short-line railroads provided connecting service to shippers located along thousands of miles of lightly used branch lines abandoned or sold by the major railroads. The major railroads had become specialists in long-haul freight. Despite large reductions in road mileage and rolling stock, their operational capacity had increased, and ton-mileage grew by more than 40 percent from 1980 to 1995. Free to control their operations, reshape their physical plant, and innovate, the railroads even reclaimed a role in the transportation of high-value freight, as they tailored their services to accommodate the long-distance movement of intermodal containers.

The statistics suggest that shippers in general benefited significantly from the more innovative and efficient postderegulation railroads that had become competitive with trucks for more traffic. Productivity gains were largely passed along to shippers through lower rates and improved service. Revenue per ton-mile had declined in real terms by more than 10 percent by the end of the 1980s, and during the next decade shippers grew accustomed to steadily declining rates as railroads continued to consolidate traffic on main lines by using larger cars and longer trains. During the 1990s, a 45 percent decrease in railroads’ real input costs was accompanied by a 30 percent drop in real rates. The few postderegulation studies that tracked and placed monetary values on changed service, such as reductions in shipper inventory costs through faster and more reliable movements, concluded that the benefits to shippers from improved rail offerings were comparable with those from the reductions in rates.

Shipper Concerns and Issues Post-2000

During the early 2000s, signs appeared that the postderegulation efficiency gains of the railroad industry were largely complete. Rail rates had begun to rise, first in nominal and then in real terms partway through the decade. Between 2002 and 2007, real rates increased by more than 15 percent. Railroads were no longer fully offsetting increases in the

price of fuel and other inputs through productivity gains. The industry’s legacy capacity excesses had been shed. Traffic volumes had nevertheless been growing through 2006, and prerecession projections of continued fast growth were causing some policy makers and shippers to express concern that capacity had become too lean. With the service disruptions that followed the merger activity during the 1990s and other service disturbances in 2004 in mind, some shippers maintained that industry consolidation and rationalization of capacity had gone too far. They expressed concern that these developments had contributed to the secular rise in rates and to a potential for future demands for rail service to go unmet.

The upward trend in rates and episodic service disruptions renewed concerns that had been expressed by some shippers since ICC’s initial implementation of the Staggers Rail Act:

• Shippers maintained that regulators interested in improving the financial position of the railroads by curtailing excess capacity had placed too little emphasis on protecting shippers from unreasonable rates and unreliable service. The claim was that because of mergers, which helped reduce the number of Class I railroads from 41 in 1979 to seven by 1999, fewer markets had effective railroad competitors. In terminating ICC in 1995 and creating STB to replace it, Congress left the new agency with the authority to approve mergers by using the public interest standard, despite suggestions by shippers that review authority be transferred to the U.S. Department of Justice (USDOJ). STB had agreed to the merger of the Union Pacific and Southern Pacific Railroads over the objections of USDOJ, which had raised concerns about losses in competition that could harm shippers and consumers. By 2000, STB had declared that an excess of uneconomic capacity was no longer a problem and that further railroad consolidation could risk competition losses. In 2001 STB issued new merger review rules that emphasized the preservation of competition. However, this interest could only be assessed as a component of a legacy public interest appraisal requiring consideration of many other factors, such as the perceived interest of the public in protecting or enlarging a railroad’s earning capabilities.

• Shippers claimed that railroads, in their dual capacity as common and contract carriers, were giving preference to the latter form of service and neglecting the former. By 2000, nearly half of all ton-miles were transported in contract carriage, and some commodities had migrated almost entirely out of common carriage. Other large segments of traditional rail traffic, including most grain and farm products, still relied on common carrier service. Shippers of these commodities maintained that during periods of temporary shortages in capacity they were more likely than contract shippers to endure significant delays and interruptions in service. They also complained that railroads were discouraging common carriage requests by providing inferior service, preconditioning service on costly upgrades in sidings and other infrastructure, and failing to deploy and make investments in the levels of capacity needed to ensure reliable service. Shippers wanted STB to introduce and enforce minimum service standards for common carriage.
• Shippers expressed concern that access to rate relief for common carrier service had been substantially restricted because of the procedures used by regulators in assessing the reasonableness of disputed rates. Rate cases applying the evidentiary standard introduced by ICC in 1985 for coal shippers could take years to adjudicate and had large minimum litigation costs that deterred rate challenges by shippers of smaller volumes over more varied routes. These shippers often had claims that were too small to justify the standard’s minimum litigation costs, and a standard designed for traffic flowing regularly in defined corridors was often inappropriate and unworkable for them. By the early 2000s, nearly all rate cases had involved shippers of coal, whose large volumes and fixed traffic lanes made the risk–reward trade-off in bringing a case more favorable. Simplified procedures introduced in the mid-1990s for lower-volume shippers of commodities such as chemicals and grain had not attracted much interest. The procedures were still viewed by these shippers as inappropriate to their circumstances, with decision criteria that were too uncertain to justify the litigation expense. Shippers in markets lacking effective competition argued that regulators had not provided a balanced implementation of the law. Railroad revenue adequacy was being safeguarded, but mainly by making rate relief practically unavailable to most rail shippers.

• Shippers claimed that railroads had become increasingly adept at obtaining and exploiting market power. They claimed that railroads set rates according to each shipper’s maximum willingness to pay at levels that were collectively yielding profits beyond those needed to achieve a revenue adequate system with normal rates of return. By 2006, three of the seven Class I railroads were declared revenue adequate by STB. The finding caused some shippers to call for a reevaluation of the regulatory program’s emphasis on promoting revenue adequacy and for the restructuring of rate reasonableness procedures to offer expanded opportunities for relief. Some shippers urged STB to make greater use of its authority to order interchange agreements and to mandate short-distance reciprocal switching to inject more competition in captive markets and reduce railroads’ market power.

Railroads disputed all of these claims, but shipper complaints gained force as railroad rates continued to rise during the early 2000s and traffic volumes strained railroad capacity. Congress originally requested this study in 2005 against this backdrop. During the following two years, industrywide average rates reached their peak, but ton-miles began to decline. The deep economic recession that began in late 2007 quieted many concerns about high rates, low service levels, and capacity shortages. They resurfaced as freight demand and real rates increased after 2010.

REVIEW OF RATES, SERVICE, AND CAPACITY ISSUES

When the U.S. Department of Transportation (USDOT) initiated this study in late 2013, the railroad industry had been experiencing renewed growth in demand, and rates had been rising along with input costs. An especially severe winter was about to create major service disturbances, particularly for coal, grain, and fertilizer shippers in the Upper Midwest. Concerns about long-range shortages in railroad capacity had abated since their prerecession peak. However, the growth of trainload movements of crude oil after 2011, much of it originating in the Upper Midwest, intensified the controversy over

whether railroads would devote substantial capacity to this new segment of contract traffic to the detriment of shippers who traditionally used common carriage.

Recent Trends in Rates (Chapter 2)

Since 2007, the railroad industry has been characterized by fluctuating rates, input costs, and demand. After real average rates per ton-mile reached their postderegulation low during 2001–2003, they increased by about 25 percent through 2013. Growth in rates was nearly twice as fast as that of ton-miles and far surpassed growth in input costs, which exhibited substantial volatility because of unstable fuel prices. The volatility may have led to higher contract rates, especially for long-term contracts renegotiated during periods of high fuel prices. Among major commodities, coal rates grew the fastest (up nearly 50 percent), followed by grain (up nearly 40 percent). The rate increases occurred even as railroads continued to consolidate traffic into larger shipments, a process that in the past had largely offset upward pressure on rates.

During the 2000s, more contract rates were negotiated as more bulk shippers shifted out of common carriage. By 2012, about three-quarters of nonexempt ton-miles moved under contract rates, compared with less than half a decade earlier. During the decade, coal shippers had turned almost exclusively to contract carriage. Shippers of grain remained the most committed to common carriage and to posted tariff rates, which still accounted for more than three-quarters of grain ton-miles in 2012.

Recent Service Quality Issues (Chapter 2)

STB maintains a waybill sampling program that allows the monitoring of railroad traffic and rates at the shipment level. It does not collect comparable shipment-specific records for monitoring the quality of common carrier service. Trends in service reliability, speed, and other aspects of performance must be inferred from a largely anecdotal record of shipper complaints. The record suggests that railroad service continues to experience disturbances when traffic volumes escalate unexpectedly and surpass the railroads’ deployment of capacity.

Service during the winter of 2013–2014 was particularly problematic for this reason and was made worse by unusually cold conditions that slowed and reduced the size of trains.

Whether service problems during such episodes are more severe for common carrier traffic or reliability is routinely inferior for that traffic cannot be ascertained from the complaint records or by assessing the aggregated service-related data collected by STB. The STB complaint record is naturally skewed toward shippers using common carriage because only their service is overseen by STB. An objective assessment of service quality might have been possible if shipment-specific data on service quality were available. In particular, whether the service provided in common carriage is substantially inferior to that of contract carriage and whether any service differentials tend to change when capacity is tight might have been ascertained.

Long-Run Supply of Capacity (Chapter 2)

Railroads maintain that service disturbances are not indicative of chronic underinvestment in capacity. Instead, they claim that disturbances are a temporary phenomenon arising from a short-run inability to adjust supply that can cause traffic to move slowly and some normally profitable traffic to go unserved. Nevertheless, concerns about railroads falling substantially short of the investments required to handle growth in freight traffic were prevalent before the recent recession. At that time, the railroad industry’s networks had been made lean, traffic had been steadily growing, and forecasts of rapid rail freight growth for the 2020–2035 time frame had become exaggerated by the 2006 ton-mile peak. The predictions of capacity gaps were often dire but were seldom accompanied by explanations of how the profit motive of railroads would allow such a suboptimal outcome to persist over periods in which adjustments could be made. A profit-maximizing railroad with access to credit markets (i.e., one that is revenue adequate) and that can price according to each customer’s willingness to pay should, in general, have the ability and incentive to invest in the capacity required to accommodate all profit-generating traffic. The profit incentive should be sufficient to motivate railroads to avoid large and protracted capacity shortfalls.

The profit motive by itself may not produce an equilibrium rail output that maximizes public welfare when externalities are considered. Forecasts of long-run capacity shortages seldom distinguish between valid concerns about railroads underinvesting in the capacity needed to handle socially optimal traffic and less credible concerns about railroads underinvesting in the capacity required to handle all profitable traffic. Shifting more freight from truck to rail may create positive externalities such as reductions in air pollution or highway congestion that neither carriers nor shippers take into account. In this sense, railroads may fall short of supplying welfare-maximizing levels of rail capacity, and this might warrant policy interventions to fill the gap. That possibility was not examined in this study because it involves issues outside the study charge and better suited to a multimodal study of national freight policy.

FINDINGS FROM A REVIEW OF RAIL REGULATION

The committee was asked to examine the effectiveness of the railroad regulatory program in balancing the interests of railroads in earning adequate returns and shippers in obtaining reasonable rate and service levels. Many aspects of the railroad regulatory program affect this balance. In view of the impracticality of reviewing them all, the study concentrated on five major regulatory provisions that remain controversial because of the substantial transformation of the railroad industry in the wake of the Staggers Rail Act.

The five provisions examined are (a) the criteria that qualify a shipper to dispute the reasonableness of a common carrier rate and the procedures used by regulators to resolve disputes, (b) the common carrier obligation and its implications for service quality, (c) the annual practice of determining the revenue adequacy of individual railroads, (d) the exemption of railroad mergers from standard antitrust reviews that focus on competition in favor of a broader public interest standard, and (e) the authority of regulators to order reciprocal switching to forestall or remedy unreasonable rates.

The results of the examination are summarized below. The chapters in which the assessments were made are identified. Key assessment findings are summarized in Box 5-1. The findings have the following

in common: many regulatory provisions and practices have outlived or no longer fulfill their original purpose and are candidates for discontinuation or replacement with practices better suited to today’s railroad industry.

Rate Relief: More Appropriate, Reliable, and Usable Procedures Are Needed (Chapter 3)

The Staggers Rail Act gave railroads substantial freedom to set prices but restricted this freedom for common carriage rates when a railroad has market dominance. Market dominance is defined in the law as the absence of effective competition from other railroads or modes of

transportation. A shipper is eligible to challenge a rate if the railroad involved has market dominance and the revenue earned from the disputed rate is more than 180 percent of the railroad’s “variable cost” of providing the service, as computed by STB. If a disputed rate qualifies for adjudication according to these criteria, STB can rule on whether the challenged rate is reasonable. If the agency finds that the rate is unreasonable, it must order the reimbursement of overcharges and may prescribe a maximum rate that can be charged.

The three main steps in STB’s granting of rate relief are (a) estimating the variable cost of a disputed movement, (b) determining whether effective competition exists in a market subject to a rate dispute, and (c) ruling on whether a qualifying rate is reasonable. The following are the committee’s findings concerning the procedures used for each step.

Variable Cost Allocations Are Invalid and Unreliable

In stipulating that the variable cost of a priced unit of traffic be calculated as a means of screening it for eligibility for rate relief, the Staggers Rail Act created an insoluble problem for regulators. Railroads produce numerous kinds of freight service that differ in many dimensions, each of which can affect pricing to a significant degree, such as volume, location, and time of travel. Characterizing the relevant product, or “unit level,” at which a shipment was actually priced can require descriptive information at a fairly precise product level. That product-specific information may be available, but cost information at the same level will not be available. This information imbalance alone is a serious problem when generalized variable cost estimates are assigned to individual shipments and are supposed to have a meaningful relationship to their prices, which are in turn product specific.

If regulators were only interested in the incremental costs of moving a given priced unit of traffic—such as the contribution to a train’s fuel use—the lack of product-specific information would be problematic in itself. However, if the interest is in trying to assign many other railroad costs to the shipment, no level of precision will suffice. Most of a railroad’s costs are shared by multiple units of traffic, not only at the aggregate level of the railroad and the network but also at the level of individual train service. For example, the sum of the incremental costs of each shipment in a multishipment train will be

less than the incremental cost of operating the train itself. The former will not include the labor costs associated with operating the train’s locomotive, which would not be reduced if any single shipment were removed. Any allocation of that common cost to the individual shipments would be arbitrary and lack an economic foundation.

In view of the futility of allocating common costs incurred at the microlevel of the train to individual shipments, making such allocations on the basis of macrolevel expense information for an entire railroad appears even less justifiable. Yet that is exactly what regulators try to do. Each category of a railroad’s annual expenses is characterized as more or less variable with respect to traffic output, as measured for a given time interval. Depending on the time interval selected—which tends to be a function of the data collection intervals—the subtotal of expenses that are considered variable with respect to traffic will differ, because some cost items can be adjusted differently over the measured period of time. For example, regulators may decide that 95 percent of fuel use is variable with respect to annual traffic output and that 50 percent of interest expenses on capital are variable. The subtotal of expenses characterized as “variable” will thus have been determined to a large degree on the basis of when and what kinds of expense data are collected. Some cost items, such as the cost of congestion and delays and the cost of risk-bearing, will be omitted. Nevertheless, regulators will divide this subtotal, including expense items as varied as fuel, locomotives, and road maintenance, and assign portions of it to segments of a railroad’s traffic. The units of traffic that will be the subject of the allocations will be fairly generic to create systemwide average variable costs for types of traffic that vaguely resemble the priced units.

After passage of the Staggers Rail Act, ICC introduced the Uniform Railroad Costing System (URCS) to fulfill the law’s requirement to estimate “the variable cost for particular movements.” URCS is a cost allocation scheme that proceeds in a manner similar to that described above. STB characterizes the results from URCS as “systemwide averages” and acknowledges that they do not reflect the actual cost of providing any specific service. Nevertheless, these averages—despite the numerous deficiencies just described—are in fact compared with the price of a specific shipment in the expectation that there will be

a meaningful relationship between the two. In view of the vagaries and inherent arbitrariness of the process, any possibility of a stable and meaningful connection appears to be low. Such a comparison is insufficient for making decisions about whether the shipment’s price is unreasonable or indicative of a railroad exercising excessive market power. Of course, URCS was developed for the primary purpose of making such decisions.

Previous studies have not found any connection between the revenue-to-variable-cost relationships that emerge from URCS and a railroad’s market power. However, they indicate that URCS produces inexplicable results that should be cause for concern. For example, 20 to 30 percent of all railroad shipments have URCS-assigned variable costs that are higher than their rates. This outcome, which has persisted in the data for at least 15 years, implies that railroads lose money on about one-quarter of their traffic. STB has defended such findings by stating that ratios below 100 percent are possible for some traffic for short time periods. The reason given is that URCS is not a measure of short-run variable costs but instead is a measure of “intermediate-run variable costs” and includes costs associated with items such as rails and ties that are fixed in the short term. That defense illustrates the inherent problem of cost allocation: how, for example, can URCS decide on the allocation of rail and tie costs to specific units of traffic? It is also an implausible defense when 20 to 30 percent of traffic is assumed to be priced below cost from year to year and railroads nevertheless remain financially solvent. URCS produces many other results that appear unreasonable on their face and that suggest systematic biases in its allocation rules. Most short-haul traffic, for example, exceeds the 180 percent revenue-to-variable-cost threshold, as do most shipments of hazardous materials. Hence, URCS makes arbitrary distinctions about which traffic qualifies for rate relief, and it does so in a nonrandom manner that is untenable.

Accordingly, the committee finds that URCS is neither an economically meaningful nor a reliable tool for making regulatory determinations about eligibility to pursue rate relief. Furthermore, the deficiencies of URCS cannot be overcome by revising it. No allocation of common costs can produce an economically valid measure of the variable cost of a shipment. When the Staggers Rail Act introduced

the variable cost formula in 1980, railroad rates had long been regulated, and data on market-based rates were not available. That void in market-based rate data no longer exists. An alternative approach to the statutory formula and to URCS-like cost allocations is to compare disputed rates with those rates paid for comparable shipments in markets that are not dominated by a single carrier. Comparison and categorization of shipments are complicated because of the aforementioned problem of identifying the appropriate product offering (that is, the “priced unit” of traffic). Traffic and rate data reported to STB cannot be used to account for all shipment characteristics that can influence pricing. However, a comparison of rates that controls for a reasonable number of observable shipment characteristics eliminates the need to assign relevant costs to the traffic, which has proved to be a futile exercise with many side effects.

As demonstrated in Appendix B, such a competitive rate benchmarking method can be implemented today because STB already maintains a detailed database of shipments whose rates are determined in the marketplace. While those data could be refined, replacement of cost allocation methods such as URCS with more credible, empirically based tools for identifying unusually high rates is now more practical.

Time Limits on Market Dominance Inquiries Are Essential

When a tariff rate exceeds the 180 percent revenue-to-variable-cost threshold and a shipper paying that rate complains, the law requires a review of the competitive structure of the market in question, often referred to as a qualitative assessment of market dominance. Such assessments are site specific and can be fact intensive. During ICC’s tenure, railroads defending a rate challenge were given wide latitude during market dominance inquiries to characterize a shipper’s rail substitution possibilities, including nontransportation options. A railroad could, for example, show that a shipper or its customers could substitute another product for the one transported by the railroad (product competition) or ship to and from other locations (geographic competition). However, to expedite rate cases, in 1998 STB prohibited evidence of product and geographic competition. Railroads have subsequently petitioned for the restoration of ICC’s approach. They claim that cur-

rent market dominance inquiries have become narrowly focused and overstate a railroad’s market power. STB has denied the requests. It reports that rate cases have proceeded at a faster pace since the 1998 ruling and subsequently eased the burden on shippers of filing and processing a rate complaint.

Nontransportation forms of competition can affect a shipper’s willingness to pay for rail transportation, perhaps significantly. But the number of factors considered in rendering a definitive assessment of a shipper’s willingness to pay could become extensive and almost open-ended. The timely processing of cases would be unlikely, and the fixed cost of challenging a rate would be raised beyond the reward potential of shippers with smaller claims. Nevertheless, antitrust agencies routinely consider product and geographic competition in defining the relevant market during merger reviews. Rather than categorically prohibiting certain evidence, they follow legislated timelines for the conduct of the reviews. Their focus is on disciplining the process directly through deadlines rather than indirectly through limits on evidentiary content.

In the same way, timelines could expedite market dominance inquiries. In so doing they would allow for an end to the categorical exclusion of types of evidence that the railroads believe is important. Indeed, strict timelines for reviews are fundamental in preventing delays in market dominance inquiries. With time limits, categorical limits on evidence are unnecessary.

Methods for Assessing Rate Reasonableness Lack a Sound Economic Rationale and Are Unusable by Most Shippers

If a shipper can prove it ships in a dominated market and its rate exceeds 180 percent of the URCS-determined variable cost, it is eligible to challenge its rate on the basis of one of three methods for judging rate reasonableness. The original method adopted by ICC in 1985 is a stand-alone cost (SAC) proceeding. In response to congressional demands for faster handling of rate cases, in 1997 STB instituted two additional methods intended for use mostly for smaller claims. All three are predicated on the idea that shippers should be required to demonstrate that the challenged rate produces revenues higher than those needed by the railroad to cover its common costs. Regulators have sought to link the decision about what constitutes a reasonable

rate to the law’s interest in ensuring that railroads are not denied the ability to achieve revenue adequacy through the imposition of such evidentiary requirements.

The SAC procedure remains STB’s main evidentiary method. It requires complainant shippers to design a hypothetical railroad that offers stand-alone service. Detailed assumptions and documentation about its configuration and investment and operating expense items such as locomotives, car leasing, personnel, materials, and administration must be provided. Rates based on stand-alone costs were initially proposed for use in the telecommunications industry, in which sunk costs are relatively low. The argument was that in telecommunication a rate that exceeds the SAC of providing the service in question could invite inefficient entry and should thus be lowered. Use of the SAC test in today’s railroad industry, in which sunk costs are massive and new railroad entry is almost inconceivable, lacks a similar rationale.

In a SAC proceeding, shippers can propose the inclusion of traffic that crosses over the corridor (or set of corridors) and contributes net revenues (profits) that effectively lower the amount of revenue the stand-alone railroad would need to earn from the complainant shipper to maintain the service. Railroads can argue that some or all of this proposed crossover traffic should be excluded. For shippers of relatively small quantities, whose traffic is not the dominant flow in the corridor, the significance of the railroad’s profits earned from crossover traffic is crucial. The profit contribution from crossover traffic is estimated by using revenue-over-variable-cost markups derived from URCS. If STB finds that the revenue earned by the defendant railroad from the complainant shipper exceeds the revenue needed by the stand-alone railroad to serve this traffic, after profits from crossover traffic are factored in, STB will find the rate to be unreasonable. The revenue-adequate rate the stand-alone railroad would need to charge the shipper would become the maximum rate that can be judged reasonable according to STB. If that rate is lower than the defendant railroad’s rate, it becomes the basis for the imposition of overcharge penalties and a prescribed rate that is no lower than 180 percent of the traffic’s revenue-to-variable-cost ratio as derived from URCS.

The SAC process was originally introduced to resolve rate disputes brought by coal shippers, who ship large, regular volumes in fixed cor-

ridors. According to coal shippers, the SAC process is burdensome and leads to rates being judged reasonable at conservatively high levels out of deference to revenue adequacy; the true revenue needs of the railroad are lower because of network economies. Coal shippers have long demonstrated the ability to bring and prevail in a SAC case. In contrast, shippers who transport smaller volumes on more varied routes have not. The design of a stand-alone railroad entails large litigation expenses that cannot be justified by shippers with relatively small claims. These shippers, whose traffic is not the dominant flow (or remotely close to it), must depend heavily on the profits generated by any crossover traffic that STB rules can be included in the SAC analysis. These profit contributions are computed from revenue-over-variable-cost markups as contrived by URCS. Complainant shippers can therefore face substantial uncertainty about a fundamental aspect of their SAC case. Such uncertainties, coupled with the high litigation costs of SAC, have discouraged most shippers of commodities other than coal from making use of the process.

After more than a decade of complaints about SAC from shippers of grain, chemicals, and other commodities, STB introduced the two expedited evidentiary methods for assessing rate reasonableness. One is a somewhat simplified version of SAC; the other, known as the three-benchmark method, assesses the profitability of traffic as determined by the revenue-to-variable-cost ratios established from URCS. In so doing, the agency indicated its continued commitment to a cost-based approach for assessing rate reasonableness and to linking the assessment process to the interest of ensuring railroad revenue adequacy. Shippers of some commodities that are heavy users of common carriage, including grain, have not demonstrated the ability to use the expedited methods after more than 15 years. They contend that the procedures continue to be irrelevant, to impose a substantial cost burden, and to be prone to uncertainties about decision criteria.

STB has been using the conceptually flawed URCS and the inappropriate SAC test for 30 years. The agency’s rate relief procedures are still not usable by many of the primary users of common carriage. The committee finds that the commitment to the SAC test and other URCS-dependent methods for assessing disputed rates has produced large and prolonged inequalities in shipper access to

the law’s maximum rate protections. The commitment to costing methods stems from congressional directives to ensure that maximum rate decisions not impair the ability of railroads to become revenue adequate. However, the revenue-to-variable-cost formula for screening rates for eligibility for regulatory review is both arbitrary and biased because of the inherent problems associated with defining and allocating variable costs. In view of these problems, it is reasonable to surmise that the SAC process was instituted as a safeguard for revenue adequacy out of concern that unjustified rate cases could put railroad financial viability at risk. The committee has found that URCS produces revenue-to-variable-cost percentages implying that one-quarter of all railroad traffic is priced below variable cost and that most hazardous materials and short-haul traffic exceeds the 180 percent threshold. These findings suggest that concerns over the reliability of the screening process are well founded. The serious deficiencies in the current screening process need to be rectified before more usable procedures for rate dispute resolution that respect the law’s interest in revenue adequacy can be implemented.

If a more reliable screening process is implemented, faster, sounder, more transparent, and more economical methods are available for resolving rate disputes that would give more shippers the opportunity to pursue rate relief. The experience in Canada has shown that time-limited arbitration, particularly if applied with a final-offer decision rule, can produce reasonably fast resolutions to rate disputes, in part by inducing settlements. STB has considered final-offer arbitration as a potentially faster and more economical means of resolving rate disputes but has concluded that such an approach would need to be legislatively authorized. The findings of this study suggest that such an authorization would be ill-advised unless it is accompanied by complementary changes in other elements of the rate relief process, especially the revenue-to-variable-cost screen.

Common Carrier Obligation: Service Quality Data Are Essential in Assuring Responsive Performance (Chapter 4)

Rail regulators historically focused their attention on rates, whose levels were often more pertinent to shippers because of the uniformity

of other attributes of regulated rail service. Until the Staggers Rail Act, all railroad traffic was moved in common carriage, and all rates and other terms of service were publicly posted and to a large degree similar. The act retained the common carrier duty but transformed both its applicability and its enforceability. The law not only ended the general applicability of common carriage but also increased the likelihood that its service attributes would become more heterogeneous. Thus, even as regulators retained the authority to establish certain common practices for aspects of common carriage such as tariff disclosure methods, rates and other attributes of the service content became more varied.

Thirty-five years after passage of the Staggers Rail Act, the broader logistics system as well as the railroads has been transformed. A requirement for common carriage that neglects service quality no longer seems tenable. At the same time, a well-defined set of service standards appears to be impractical in an environment characterized by diverse service offerings and changing logistics needs. The dilemma for regulators is that for the common carrier obligation to persist—as it must, if only to give effect to the law’s protections for shippers from unreasonable rates—the capability of monitoring service levels is essential in ensuring that the obligation is being met.

The tracking of complaint reports is not adequate for this purpose, nor are highly aggregated data that do not even distinguish between the service that is being provided for common and contract carriage. STB recognizes the need for better service-related data. However, its proposed enhancements appear to be an ad hoc reaction to the latest episode of service disturbances and complaints. The committee finds that regularly collected, usable data on service quality are needed to evaluate service performance. In particular, shipment-specific data could help ascertain whether service provided in common carriage is substantially inferior to that provided in contract carriage and whether any service differentials change markedly when capacity is tight. The waybill sampling program that STB uses for monitoring railroad traffic and rates is one model for shipment-level data, and examples can be found in other transportation industries, such as the on-time performance data that are collected for each airline flight.

Annual Revenue Adequacy Determination Serves No Constructive Purpose (Chapter 4)

On an annual basis, STB compares each Class I railroad’s rate of return on investment with an estimate of the industrywide cost of capital. Railroads with a rate of return exceeding the cost of capital are declared revenue adequate. The requirement was introduced during the 1970s, when many railroads were on the edge of bankruptcy and some were receiving substantial government subsidies. Policy makers needed to gauge the effectiveness of the regulatory reforms in rescuing the industry from its financial distress. That concern no longer exists.

There is a continuing public interest in ensuring the ability of railroads to keep investing, but balanced with an interest in preventing the excessive exercise of market power. In essence, policy makers need to determine whether a railroad’s profits are consistently outside a reasonable band of profitability that characterizes many other industries over a business cycle. This need is not met by a regulatory agency annually issuing a single assessment of each railroad’s earnings performance. Like all businesses, railroads have good and bad financial periods. A process that boils down financial and economic performance into a single pass/fail judgment is misleading and incapable of providing the detail needed for informed policy making. Furthermore, by continuing to compare each railroad’s rate of return with an industrywide average cost of capital, regulators leave the impression that the practice might eventually be used to impose rate-of-return regulation. Such public utility–type regulation has never been used to regulate railroads and would be at odds with the Staggers Rail Act, a central policy of which is to minimize the need for federal regulatory control.

The committee finds that the annual revenue adequacy determination no longer provides meaningful information for policy making. Its persistence prolongs the misguided view that a single yes/no indicator of railroad profitability can be used to regulate rates. Periodic, but not annual, reviews of the economic and financial condition of the railroad industry as a whole, including rate, service, and competition levels, would provide a richer set of information for the improvement of railroad policies and regulation.

Merger Review: The Legacy Public Interest Standard Is No Longer Justified (Chapter 4)

Two Class I railroads seeking to merge must apply and obtain approval from STB, which inherited this review authority from ICC. By law, STB must consider a range of potential effects from a merger, including impacts on the competitive structure of markets, rail workers, safety, and the ability of the applicants and other railroads to earn adequate revenues. In contrast, merger reviews conducted by the antitrust agencies focus exclusively on whether the transaction is likely to “substantially lessen competition.” In most transportation industries, merger reviews are conducted by the Antitrust Division of USDOJ on the basis of well-defined and transparent analytic methods, evidentiary procedures, and review timelines.

After the number of Class I railroads had declined to seven by the end of the 1990s, STB expressed concern that further consolidation would harm competition to the detriment of consumers. In 2001 STB introduced new procedures for reviewing mergers that placed greater emphasis on preserving competition. Because the law itself was not changed, the agency remained obligated to appraise mergers in the context of the broader, statutorily required public interest standard. That standard allows ill-defined and ambiguous interests to be introduced into the review. It also allows consideration of potential outcomes that are no longer clearly in the public interest, such as the ability of the transaction to raise the revenues of the merged railroad and to protect the earnings capability of other railroads competing with the merged railroad. When most railroads were financially troubled, giving positive consideration to such effects may have been in the public interest, if only to reduce the need for public subsidy. However, such conditions no longer exist.

In antitrust merger reviews, the enforcement agency must prove that the transaction is illegal because it is likely to lessen competition substantially. Under STB merger review procedures, the railroad applicants must demonstrate that losses in competition will be minimal or that such losses will be more than offset by other beneficial outcomes in the public interest. The committee’s assessment of STB’s merger review procedures could find no definitive guidance on what

constitutes a merger outcome that would or would not be in the public interest, nor any clear directives about the evidentiary and analytic procedures that applicants must follow to inform the agency’s decision. This clouded review process appears to have arisen from a statutory requirement to review mergers in the context of a broader standard. The standard has been applied in the past to reduce the uneconomic capacity of struggling railroads and to concentrate traffic and revenues for the financially healthier railroads that remained. Railroad financial stability has been achieved, and any further merger reviews are likely to hinge on efficiency and competition issues that USDOJ is most qualified to assess. Thus, the rationale for retaining STB’s role in reviewing mergers according to a public interest standard is no longer compelling.

The committee finds that there is no longer an economically sound argument for retaining the ambiguous public interest standard applied by regulators for merger reviews in lieu of a well-defined competition-based appraisal by antitrust enforcers. The law already requires USDOJ to advise STB on the competition impacts of major railroad mergers. USDOJ’s Antitrust Division is more qualified to assess competitive impacts than is STB, and the rationale for relegating the antitrust agency to a subordinate role no longer exists.

Reciprocal Switching Orders: A Potential Remedy for Unreasonable Rates (Chapter 4)

The Staggers Rail Act made it easier for a railroad to cancel terminal access, trackage rights, and reciprocal switching agreements with competitors for traffic that it could serve by itself. The law nevertheless gave regulators a new authority for ordering reciprocal switching arrangements when that was “necessary to provide competitive rail service.” A railroad ordered to engage in reciprocal switching must agree, for a regulated fee, to transport shipments originating or terminating on its line to or from a nearby interchange with another railroad willing to perform the line-haul movement. In this way, a reciprocal switching order can increase a captive shipper’s competitive options. Canada’s two major railroads have long been required to switch traffic according to government-prescribed switching rates if requested for

an interchange located within 30 kilometers (about 19 miles) of the shipment’s origin or destination.

Reciprocal switching has been ordered on rare occasions in the United States to address competitive abuses that led to inadequate service. ICC’s rationale for making minimal use of the authority, to which STB adheres, is that the law’s maximum rate provision is available to captive shippers to obtain reasonable rates. In addition, reciprocal switching has not been imposed as a remedy for rates ruled unreasonable because the law expressly allows railroads to price up to 180 percent of variable cost, and a reciprocal switching order could depress rates below that level. The experience in Canada, where reciprocal switching is required regardless of rate levels, provides an opportunity for assessing its effects, particularly if the interest is in broader application. One possible starting point for assessing reciprocal switching on a more limited basis is to allow its use as an optional remedy for rates that have been ruled unreasonable and thereby to provide an alternative to a prescribed rate.

RECOMMENDATIONS FOR REGULATORY CHANGE

A presumption of this study has been that policy makers are satisfied with the overarching policies of the Staggers Rail Act, such as allowing railroads to achieve revenue adequacy and shippers to obtain reasonable rates. The study findings suggest that certain regulatory provisions and practices no longer serve these policy goals or could serve them more effectively if they were designed and implemented differently. Hence, most of the recommendations that follow do not simply advise ending a regulatory provision or practice; they are accompanied by proposals for alternative regulatory designs or procedures better suited to today’s railroad industry. Because the study was called for in legislation, the intended main recipient of the recommendations is Congress. Most of the recommended steps summarized in Box 5-2 would require legislative action.

Prepare to repeal the 180 percent revenue-to-variable-cost formula by directing USDOT to develop, test, and refine competitive rate benchmarking methods that can replace URCS in screening rates for eligibility to be challenged.

Recommendation: End the economically unsound regulatory requirement of estimating the variable cost of railroad movements to allow STB to dispense with all cost allocation schemes such as URCS. Move closer to common law notions of rate fairness by introducing a rate screening process that allows shippers to seek relief if they are paying tariff rates that are unusually high in comparison with rates paid for similar shipments in markets having more competition options. Require USDOT to make a concerted effort to develop a competitive rate benchmarking system to replace URCS and the revenue-to-variable-cost formula, and make the necessary resources for this task available.

Rationale: When the revenue-to-variable-cost formula was introduced in the Staggers Rail Act more than 30 years ago, rates had long been set by cartels under regulatory oversight. There was little prac-

tical value in comparing rates in dominated versus more competitive markets. The regulatory practice of assigning portions of a railroad’s total costs to individual units of traffic was therefore retained as a way of identifying potentially excessive rates. However, most of a railroad’s costs are shared by multiple units of traffic and cannot be unambiguously divided and assigned to individually priced units. The variable cost estimations of URCS are arbitrary and bear no stable relationship to the cost of individual shipments, railroad pricing decisions, or railroad revenue needs. Even though the variable cost allocations are inherently arbitrary, they play a significant role in deciding which rates qualify for relief. Comparisons with actual shipment prices are bound to produce distorted depictions of market dominance and other illogical outcomes, such as consistently showing large amounts of traffic moving at an economic loss. URCS has been shown to produce such distortions.

A wealth of information on unregulated, market-based rail prices now exists in STB’s annual Carload Waybill Sample (CWS), along with detailed information on characteristics of individual shipments. The information can be used to develop models of rates determined under effectively competitive conditions. The resulting models can be used by a shipper to compare its rate with the model’s benchmark prediction of the rate it would have been charged under effective competition. Even if a railroad’s rates were uninfluenced by levels of competition, about half the time shippers who compared their rates with the median predicted benchmark would find their rates to be higher, at least to some degree. No model based on real data ever fits perfectly, and prediction errors will cause many tested rates to exceed their predicted competitive benchmark levels. Sometimes the excess will be large in percentage terms. The reason is that the data cannot include all of the economically meaningful characteristics of a shipment that may affect its rate. However, the larger the excess, the greater the likelihood that unobserved characteristics and prediction error alone are not the cause and that the exercise of market power is a contributing factor.

Regulators would need to decide the percentage by which a tested rate must exceed its predicted benchmark competitive level before a shipper could have its rate scrutinized as potentially unreasonable. Examinations of each model’s predictive capability may be undertaken

to inform such decisions, and regulators may apply different qualifying thresholds for different commodities. Regulators would need to consider an obvious trade-off in making such decisions: a stricter screen (i.e., large percentage threshold) will provide less of a threat to railroad revenue adequacy, but fewer shippers with legitimate rate grievances will be eligible for relief. Making the screen less strict will offer greater opportunity for shippers to challenge their rates but pose a greater threat to railroad revenue adequacy. Decisions about the appropriate threshold could be controversial. However, the transparency of the competitive rate benchmarking tool makes it preferable to the current system, which relies on arbitrary cost allocation rules that are used to implement an arbitrary revenue-to-variable-cost formula contrived more than a generation ago.

The recommended competitive rate benchmarking model approach would be no more complicated to construct and run, and would probably be less so, than the annual derivation of variable costs from URCS. The complexity of the latter has prevented its basic structure from being changed for decades despite fundamental methodological flaws. The proof-of-concept analysis in Appendix B demonstrates how the CWS, supplemented with readily available information on the physical layout of the railroad networks and proximity to competing transportation modes, can be used to develop the benchmark models. Once they were constructed, models of this general type could be used from year to year. They would be updated regularly with new data on shipment rates and characteristics as obtained from the annual CWS. Refraining from making significant changes in the models except on a periodic basis (e.g., every 5 to 10 years) would have the advantage of ensuring stability and predictability by users.

Make entitlement to rate relief contingent on a satisfactory finding of market dominance in the procedure for ruling on the reasonableness of a rate and do not limit the types of evidence that can be used to assess dominance.

Recommendation: Make a shipper’s entitlement to rate relief dependent on a satisfactory finding of market dominance in the procedure used for determining the reasonableness of the rate to avoid delays in the process-

ing of cases. There should be no restrictions on the types of evidence—such as that pertaining to product and geographic competition—that can be introduced to assess market dominance.

Rationale: The law does not clearly define what constitutes an unreasonable rate, but it does stipulate that relief should apply when there is a lack of effective competition. A competitive rate benchmarking model cannot control for all factors that can temper a railroad’s market power, including product and geographic competition. Thus, further assurance is warranted that a railroad’s market dominance has been directly reviewed before a shipper is entitled to relief. The structure and elements of that dominance review need not be formally defined and can be folded into the overall process for ruling on the reasonableness of a rate as long as the design of the overall process prompts timely rulings.

STB’s current detailed assessment of market dominance deters shipper claims by delaying processing and increasing litigation costs. The process restricts the evidence a railroad can submit to dispute market dominance as a way of expediting this step. The emphasis on expedited processing should remain, but timeliness should be achieved by integrating the assessment of market dominance into the procedure used for ruling on the reasonableness of the rate. The antitrust agencies now consider evidence on market structure under a strict timeline during complex merger reviews, and there is no obvious reason why the assessment of market dominance for a less complicated rail rate dispute should require a separate, time-consuming proceeding.

Replace STB rate reasonableness hearings with an arbitration procedure that compels faster rulings on disputes involving rates found eligible to be challenged because they substantially exceed their competitive benchmarks.

Recommendation: End STB’s direct role in adjudicating rate disputes and prescribing penalties and remedies. Require that disputes involving rates that pass the benchmarking screen be resolved through an independent arbitration process similar to the one long used for resolving rate disputes in Canada. Unless both sides agree to another

format, the arbitration should be performed under a strict time limit and a final-offer rule, whereby each side offers its evidence, arguments, and possibly a changed rate or other remedy in a complete and unmodifiable form after a brief hearing. The arbitrator should be instructed to keep the offers private and choose only one side’s full offer without compromise. However, if market dominance is not found to the satisfaction of the arbitrator, he or she should be guided by STB instructions either to dismiss the challenge or to accept the railroad’s offer. STB should identify candidate arbitrators and require professional qualifications that are not so restrictive with regard to specialized railroad industry expertise that the processing of challenges may be slowed. Serious consideration should be given to restricting opportunities to appeal arbitration rulings to ensure that the process remains timely and economical.

Rationale: The evidentiary standards and procedures used by ICC and STB for adjudicating rate disputes are slow, costly, and inappropriate to many shippers’ circumstances. They prevent shippers from having equal and effective access to the law’s maximum rate protections. Efforts to streamline and expedite the procedures have not overcome these deficiencies. In some respects they have made matters worse by causing STB to become more dependent on the arbitrary cost allocations made by URCS. Thus, STB has moved toward replacing the inappropriate and cumbersome SAC test with procedures that offer even less predictable decision criteria and lack even that test’s weak conceptual basis.

The recommended replacement of formal STB hearings with a private, final-offer arbitration process would motivate the sides to pursue constructive resolutions without concern over establishing unfavorable precedents. Arbitration accompanied by deadlines should make the processing of rate disputes faster and induce settlements. The stipulation that the briefs, final offers, and explanations for the decision are to be kept confidential (and thus would not be precedent setting) should increase the chances of a settlement. The ruling must be consistent with the law’s interest in controlling rates only when there is market dominance, so the arbitrator must be convinced of dominance either

as a condition for proceeding to the final-offer stage or for choosing the shipper’s final offer. The first approach, requiring a finding of market dominance as a prerequisite for reviewing final offers, may be viewed as disadvantageous by a shipper with a smaller claim; the prerequisite could introduce a time-consuming step before a final ruling can be made in the shipper’s favor. The latter approach, which implies a single-stage process, may be viewed as disadvantageous by a railroad. It could be put in the position of having to choose how far to go in disputing market dominance versus defending the reasonableness of its rate or proposing alternatives in its final offer. The timing of the introduction of the market dominance decision into the arbitration proceeding would need to take such views into account. Presumably, other arbitration formats and dispute resolution methods could be used if they were agreed to by all parties. Among them might be a separate arbitration to decide first on the reasonableness of the rate and, if it is found to be unreasonable, a second stage to decide the remedy.

In the committee’s view, access to rate relief on the basis of the competitive rate benchmarking system would end the necessity for elaborate evidentiary procedures such as SAC presentations. The URCS-based revenue-to-variable-cost formula is an unreliable indicator of the exercise of market power. It offers no assurance that rates above its threshold are unreasonable or that rates below are not. STB has imposed burdensome evidentiary standards for rate adjudication, perhaps to protect revenue adequacy. In addition to being inappropriate for many types of traffic, these standards have such high fixed litigation costs or uncertain decision criteria that they have substantially restricted the pool of shippers eligible to pursue rate relief. Under the proposed competitive rate benchmarking approach, revenue adequacy could be protected at the outset of the process on the basis of transparent choices made by regulators about the strictness of the benchmarking screens.

Use of the proposed competitive rate benchmarking tool would allow all subsequent phases of the rate relief process to be made more economical and usable, since they would no longer serve as the main safeguards for revenue adequacy. The committee believes that if a competitive rate benchmarking process is instituted, STB’s current adjudication processes can be replaced by the expedited form of rate

arbitration long used in Canada without threatening revenue adequacy. The strict timeline for preparing arguments and evidence should make the process more economical and thus more accessible to shippers who pass the competitive rate and market dominance screens. It should also prompt more shippers and railroads to resolve disputes on their own.

Allow reciprocal switching as a remedy for unreasonable rates.

Recommendation: End the practical prohibition on reciprocal switching as a remedy for an unreasonable rate. Allow the parties in rate arbitrations to propose reciprocal switching arrangements in their offers to resolve the dispute if they so desire and allow the arbitrator to order that such arrangements be made.

Rationale: Allowing a reciprocal switching arrangement to be proposed as a remedy in a final-offer arbitration proceeding appears to be prudent. Any proposal under this format, if the proposer intends to prevail, is likely to be reasonable in scope. Accordingly, there should be no need for regulators to set switching fee schedules or to establish applicable distance limits as in Canada. Such terms should be part of the offer to be put before the arbitrator. Reciprocal switching has never been prescribed by STB when a rate is found to be unreasonable, partly out of concern that such an intervention would cause rates to go below the statutory 180 percent revenue-to-variable-cost threshold. The recommended repeal of this arbitrary formula should make this concern moot.

End annual revenue adequacy determinations; require periodic assessments of industrywide economic and competitive conditions.

Recommendation: End the requirement for annual determinations of each railroad’s revenue adequacy status. Replace this formulaic process with a requirement for periodic (e.g., 5-year) monitoring and assessment of the railroad industry’s economic performance, competitive conditions, and rate and service levels to inform railroad regulatory decisions and policies.

Rationale: The annual pass/fail revenue adequacy appraisal of Class I railroads has become ritualistic. It offers little substantive information for regulators and policy makers in monitoring economic and competitive conditions in the industry. The requirement was adopted when railroads were failing and the subject of government rescue efforts. It suggests a long-term interest in regulating the profitability of individual railroads, which appears neither practical nor consistent with the deregulatory thrust of the Staggers Rail Act reforms. By sponsoring periodic assessments of economic and competitive conditions in the industry as a whole on the basis of more varied data and analytic techniques, Congress and STB would obtain a richer set of information to support regulatory decisions and policies.

Transfer merger review authority to the antitrust agencies, which would apply customary antitrust principles rather than a public interest standard.

Recommendation: End the requirement that railroad mergers be reviewed and approved by STB according to a broad public interest standard. Turn over responsibility for merger reviews to USDOJ, which would enforce the competition standard of the antitrust laws as it is applied to other transportation industries. Just as it consults with USDOT on airline mergers, USDOJ would be expected to solicit the views of STB on proposed railroad mergers, but it would not be bound by those views.

Rationale: Decades ago, when the railroads were heavily regulated, they were exempted from antitrust reviews of proposed mergers and subjected instead to a broader public interest review by ICC. Even after economic regulations in the industry were eased, the public interest standard was retained. Part of the purpose was to allow the more financially viable railroads to reduce perceived duplicative capacity by acquiring struggling competitors. In that way, they could concentrate traffic and revenues and regain profitability. Any such rationale for keeping the public interest standard no longer exists. STB itself has stated that excess and duplicative capacity are no longer problems and that preserving competition among the remaining railroads will be the

priority for future reviews. In view of the diminished reasons for the public interest standard, its preservation can only detract from the desired focus on competition. STB is not as qualified to assess competitive effects as the Antitrust Division of USDOJ, which is already required to advise STB on a merger’s potential competition effects. Ending the public interest standard and turning over responsibility to the Antitrust Division, with an obligation to hear STB’s views, would be timely and is warranted.

Require a strategic review of STB data programs to identify opportunities to simplify or discontinue the reporting of little-used data as a general matter and for the following specific purposes:

1. To improve the accuracy, utility, timeliness, and availability of the CWS to implement the competitive rate benchmarking tool and enable more independent analyses and beneficial uses.

Recommendation: STB should be given the direction and resources to undertake a strategic review of all of its data programs. The review should begin with the role of the CWS in enabling implementation of the recommended competitive rate benchmarking system and in facilitating academic and other research on the railroad industry that can inform policy making. The CWS is STB’s main empirically derived tool for monitoring industry traffic and revenues to support regulatory decisions, and it should be recognized as the linchpin of the agency’s data program. An effort should be made to improve its accuracy, utility, timeliness, and availability. The strategic review should recommend modifications and enhancements of the CWS that will be needed in implementing the competitive rate benchmarking tool and should identify elements of the database that can be eliminated because they are no longer used.

The review should also make recommendations on how to expand access to the CWS. They should be based on examinations not only of how confidentiality restrictions can be modified to protect business interests but also of how beneficial use of the data can be promoted by making public- and government-use versions more accurate. The latter can be particularly important in informing public investment

decisions, such as whether and where to improve the nation’s system of waterways as determined by the U.S. Army Corps of Engineers. The Bureau of the Census has long made confidential data available to researchers in electronic form with safeguards that protect confidentiality. The strategic review should examine such safeguards and those used by other federal agencies and should recommend modified versions of them suitable for the CWS.

2. To obtain shipment-level data on service quality to monitor the railroads’ responsiveness to the common carrier obligation.

Recommendation: As part of a strategic review of its data programs, STB should appraise the data needed to fulfill its role in supervising the supply of common carrier service. For example, consideration should be given to collecting information that permits the tracking of the time elapsed from a shipper request for service to rail car placement, removal, and arrival at the destination, perhaps in conjunction with information on the scheduled delivery time. The appropriate platform for such data collection may be the CWS, because shipment-level tracking of service is essential for understanding trends in service levels and patterns across time, regions, and traffic segments. STB should explore options for collecting shipment-level data, including additions to and enhancements of the CWS itself. STB should examine all data elements in the standard railroad freight waybill that could be useful for monitoring service performance and consider adding such elements to the CWS. STB should also examine opportunities for collecting new data, which would either be added to the waybill reporting or subsequently linked to CWS records.

3. To reassess the collection of detailed railroad accounting, financial, and operations data, with consideration of opportunities to reduce railroad reporting burdens as regulatory approaches change and to obtain the kinds of data that will be required to conduct periodic economic and competition studies of the industry.

Recommendation: A strategic review of STB’s data programs would not be complete without an examination of the purpose of the agency

in requiring the use of its own railroad accounting system and the reporting of railroad financial and operations information, often in copious detail. The changes in the regulatory program recommended in this report would enable STB to review these data reporting requirements and to assess how changes in them would affect its ability to discharge its remaining regulatory and oversight responsibilities. Adoption of the recommended actions, such as discontinuation of URCS, SAC assessments, public interest appraisals of mergers, and formulaic annual revenue adequacy determinations, should have far-reaching implications for the agency’s needs for railroad financial and operations data. The resources saved from any streamlining or simplification of these data programs could be used to enhance the agency’s other data programs on rates and service quality. STB could focus on the collection of the kinds of financial and economic data needed in supporting the recommended periodic economic and competition studies.

CONCLUDING COMMENTS

STB could take early steps to advance the recommendations of this report, such as by supporting USDOT in exploring competitive rate benchmarking methods and by commencing planning for a modernized data program. Such efforts could help inform the legislative actions that will likely be required to further the recommendations—actions the committee believes are overdue. The last major revision to the Staggers Rail Act terminated ICC and created STB 20 years ago. The Staggers Rail Act itself was passed 35 years ago. In the interval, the railroad industry has been transformed, essentially modernized in step with the other transportation industries that were deregulated at about the same time. The railroad industry was in a fundamentally different position at the time of its deregulation—on the edge of bankruptcy, despite its considerable potential market power, and in need of specialized regulatory reforms that took its financial distress into account. The industry continues to have characteristics differing from those of the other transportation modes, such as its vertical integration and the ability to obtain and exercise local market power, that require ongoing regulatory oversight. Therefore, railroad deregulation should not

be complete. The economic regulations that remain should be suited to the financially sound railroad industry of today, not to the foundering one that required rescue 35 years ago. The actions recommended in this report recognize the continued significance of the railroad regulatory program and are intended to resynchronize key elements of it that have become outdated.

The modernization proposed in this report would reduce the anachronistic regulatory burdens railroads still bear while giving more shippers real protection against unreasonable rates. It would continue the process begun by the Staggers Rail Act—a process aimed at producing a modern, efficient, and competitive railroad industry able to attract capital, maintain and expand its capacity, and serve its customers with a minimum of regulatory oversight.