American Commercial Lines, Inc. v. Louisville & Nashville Railroad , 88 S. Ct. 2105 ( 1968 )

Mr. Justice Marshall

delivered the opinion of the Court.

The basic issue in these cases is whether the action of the Interstate Commerce Commission in disallowing a rate reduction proposed by the appellee railroads, 326 I. C. C. 77 (1965), was consistent with the provisions of § 15a (3) of the Interstate Commerce Act, 49 U. S. C. § 15a (3), added by 72 Stat. 572 (1958), which governs ratemaking in situations involving intermodal competition. A subsidiary but related issue is whether the Commission adequately articulated its reasons for disallowing the proposed rate. A statutory three-judge court, upon appeal of the Commission’s decision by the appellee railroads, held that the Commission’s decision was erroneous on both of the foregoing grounds. 268 F. Supp. 71 (D. C. W. D. Ky. 1967). Because of the importance of § 15a (3) as the primary guide to ICC resolution of rate controversies involving intermodal competition, we noted probable jurisdiction of the appeal taken by the Commission and the competing carriers from the decision of the District Court.¹ 389 U. S. 1032 (1968). For the reasons detailed below, we conclude that the District Court erred in its rejection of the Commission’s decision, and the grounds on which it was based, and we reverse.

I.

Since 1953 the movement of ingot molds from Neville Island and Pittsburgh, Pennsylvania, to Steelton, Kentucky, has been almost exclusively by combination barge-truck service, and since 1960 the overall charge for this service has been $5.11 per ton. In 1963 the Pennsylvania Railroad and the Louisville & Nashville Railroad lowered their joint rate for this same traffic from $11.86 to $5.11 per ton. The competing barge lines, joined by intervening trucking interests, protested to the ICC that the new railroad rate violated § 15a (3) of the Interstate Commerce Act because it impaired or destroyed the “inherent advantage” ² then enjoyed by the barge-truck service. The Commission thereupon undertook an investigation of the rate reduction.

In the course of the administrative proceedings that followed, the ICC made the following factual findings about which there is no real dispute among the parties. The fully distributed cost³ to the railroads of this service was $7.59 per ton, and the “long term out-of-pocket costs”⁴ were $4.69 per ton. The fully distributed cost to the barge-truck service⁵ was $5.19 per ton.⁶ The out-of-pocket cost⁷ of the barge-truck service was not separately computed, but was estimated, without contradiction, to be approximately the same as the fully distributed cost and higher, in any event, than the out-of-pocket cost of the railroads. The uncontroverted shipper testimony was to the effect that price was virtually the sole determinant of which service would be utilized, but that, were the rates charged by the railroads and the barge-truck combination the same, all the traffic would go to the railroads.

The railroads contended that they should be permitted to maintain the $5.11 rate, once it was shown to exceed the out-of-pocket cost attributable to the service, on the ground that any rate so set would enable them to make a profit on the traffic. The railroads further contended that the fact that the rate was substantially below their fully distributed cost for the service was irrelevant, since that cost in no way reflected the profitability of the traffic to them. The barge-truck interests, on the other hand, took the position that § 15a (3) required the Commission to look to the railroads' fully distributed costs in order to ascertain which of the competing modes had the inherent cost advantage on the traffic at issue. They argued that the fact that the railroads’ rate would be profitable was merely the minimum requirement under the statute. The railroads in response contended that inherent advantage should be determined by a comparison of out-of-pocket rather than fully distributed costs, and they produced several economists to testify that, from the standpoint of economic theory, the comparison of out-of-pocket, or incremental, costs was the only rational way of regulating competitive rates.

The ICC rejected the railroads’ contention that out-of-pocket costs should be the basis on which inherent advantage should be determined. The Commission observed that it had in the past regularly viewed fully distributed costs as the appropriate basis for determining which of two competing modes was the lower cost mode as regards particular traffic. It further indicated that the legislative history of § 15a (3) revealed that Congress had in mind a comparison of fully distributed costs when it inserted the reference to the National Transportation Policy into that section in place of language sought by the railroads. The Commission also emphasized that there was a rulemaking proceeding pending before it in which the whole question of the proper standard of costing in situations involving intermodal competition was being examined in depth, and stated that “a radical departure from the fully distributed cost norm” would not be justified on the basis of the record before it in this case.

Having decided to utilize a comparison between fully distributed costs to determine inherent advantage, the Commission then concluded that the rate set by the railroads would undercut the barge-truck combination’s ability to exploit its inherent advantage because the rate would force the competing carriers to go well below their own fully distributed costs to recapture the traffic from the railroads. Moreover, since the result sought by the railroads was general permission to set rates on an out-of-pocket basis, the Commission concluded that eventually the railroads could take all the traffic away from the barge-truck combination because the out-of-pocket costs of the former were lower than those of the latter and, therefore, in any rate war the railroads would be able to outlast their competitors. Accordingly, the Commission ordered that the railroads’ rate be canceled.

The District Court read the statute and its accompanying legislative history to reflect a congressional judgment that inherent advantage should be determined in most cases by a comparison of out-of-pocket costs and that, therefore, railroads should generally be permitted to set any individual rate they choose as long as that rate is compensatory.⁸ The court also held that the Commission had failed adequately to articulate its reasons for deciding that the proper way of determining which mode of transportation was the more efficient was by comparison of fully distributed costs rather than out-of-pocket costs. Although this latter holding appears first in its opinion, it is evident that it must logically follow its ruling on the meaning of § 15a (3), since if Congress in enacting that section had already decided that inherent advantage should be determined by reference to fully distributed costs, there would be no special burden on the Commission to justify its use of them.

h-l

This Court has previously had occasion to consider the meaning and legislative history of § 15a (3) of the Interstate Commerce Act in ICC v. New York, N. H. & H. R. Co., 372 U. S. 744 (1963) (“New Haven”), and both the ICC and the District Court have relied heavily on that decision as support for the conflicting results reached by them in these cases. Because the statute and its relevant legislative history were so thoroughly canvassed there, we shall not undertake any extended discussion of the same material here. Instead, we shall refer to that opinion for most of the relevant history.

So far as relevant here, § 15a (3) provides that:

“[r]ates of a carrier shall not be held up to a particular level to protect the traffic of any other mode of transportation, giving due consideration to the objectives of the national transportation policy declared in this Act.”

The National Transportation Policy, 49 U. S. C. preceding § 1, states that it is the intention of the Congress:

“to provide for fair and impartial regulation of all modes of transportation subject to the provisions of this act, so administered as to recognize and preserve the inherent advantages of each . . . .”

The enactment of § 15a (3) in 1958 was due primarily to complaints by the railroads that the ICC had maintained rates at artificially high levels in order to protect competing modes from being driven out of business by railroad competition.⁹ The bill that eventuated in the language that is presently § 15a (3) originally provided that the ICC, in considering rate reductions, should, in a proceeding involving competition with another mode of transportation, “consider the facts and circumstances attending the movement of the traffic by railroad and not by such other mode.” (Emphasis added.) 372 U. S., at 754. This language was objected to strongly by both the ICC and representatives of those carriers with which the railroads were in competition. See Hearings on S. 3778 before the Senate Committee on Interstate and Foreign Commerce, 85th Cong., 2d Sess. (1958). The basic ground of objection was that by looking only to the effect of a rate reduction on the carrier proposing it, the ICC would be unable to protect the “inherent advantages” enjoyed by competing carriers on the traffic to which a rate reduction was to be applied.

Unfortunately, the meaning of the term “inherent advantage,” which is what the Commission is supposed to protect, is nowhere spelled out in the statute. The railroads argue, and the District Court held, that Congress intended by the term to refer to situations in which one carrier could transport goods at a lower incremental cost than another. The fallacy of this argument is that it renders the term “inherent advantage” essentially meaningless in the context of the language and history of § 15a (3).

Since the pricing of railroad services below out-of-pocket or incremental cost would result in a net revenue loss to the railroad on the carriage, the ICC could prohibit such practices without reference to the costs of any other competing carrier. And this is precisely what the language of the bill as originally endorsed by the railroads would have provided by its use of the phrase “and not by such other mode.” See supra, at 580. This language was, however, rejected by the Congress and the alternative formulation proposed by the ICC, see Hearings, supra, at 169, was substituted for it.

As this Court said in the New Haven case:

“The principal reason for this reference [to the National Transportation Policy] . . . was to emphasize the power of the Commission to prevent the railroads from destroying or impairing the inherent advantages of other modes. And the precise example given to the Senate Committee, which led to the language adopted, was a case in which the railroads, by establishing on a part of their operations a compensatory rate below their fully distributed cost, forced a smaller competing lower cost mode to go below its own fully distributed cost and thus perhaps to go out of business.” 372 U. S., at 758.

Since these cases are identical to the example just described, it would seem that, at the very least, the result reached by the Commission here is presumptively in accord with the language of the statute and with the intent of Congress in utilizing that language.¹⁰

The District Court, however, ignored the above portion of the New Haven opinion and seized on certain other language therein to the effect that:

“It may be, for example, that neither a comparison of ‘out-of-pocket’ nor a comparison of ‘fully distributed’ costs, as those terms are defined by the Commission, is the appropriate method of deciding which of two competing modes has the cost advantage on a given movement.” 372 U. S., at 760.

It coupled this language with its interpretation of § 15a (3) as having the purpose to promote “hard competition,” and concluded that the Commission had the burden of justifying any departure from using out-of-pocket cost as the means of determining inherent cost advantage.

We think that the District Court erred in its reading both of the prior New Haven decision and of the extent to which Congress intended to foster intermodal competition. We note first that nothing in the language of the New Haven opinion indicates a preference for either out-of-pocket or fully distributed costs as a measure of inherent advantage; rather, all that is said is that the appropriate measure “may be” neither. Given the fact that the insertion of the reference to inherent advantage into § 15a (3) came about at the insistence of carriers that were demanding that fully distributed costs be the sole measure of that advantage,¹¹ we think that the clear import of the foregoing statement in the New Haven opinion was that the Commission could, after due consideration, decide that some other measure of comparative costs might be more satisfactory in situations involving intermodal competition than the one it had traditionally utilized.¹² That is a far cry from saying that it must.

The District Court apparently believed that the Commission was required to exercise its judgment in the direction of using out-of-pocket costs as the rate floor because that would encourage “hard”¹³ competition. We do not deny that the competition that would result from such a decision would probably be “hard.” Indeed, from the admittedly scanty evidence in this record, one might well conclude that the competition resulting from out-of-pocket ratemaking by the railroads would be so hard as to run a considerable number of presently existing barge and truck lines out of business.

We disagree, however, with the District Court’s reading of congressional intent. The language contained in § 15a (3) was the product of a bitter struggle between the railroads and their competitors. One of the specific fears of those competitors that prompted the change from the original language used in the bill was that the bill as it then read would permit essentially unregulated competition between all the various transportation modes. It was argued with considerable force that permitting the railroads to price on an out-of-pocket basis to meet competition would result in the eventual complete triumph of the railroads in inter-modal competition because of their ability to impose all their constant costs ¹⁴ on traffic for which there was no competition.

The economists who testified for the railroads in this case all stated that such an unequal allocation of constant costs among shippers on the basis of demand for railroad service, i. e., on the existence of competition for particular traffic,¹⁵ was economically sound and desirable. Apart from the merits of this contention as a matter of economic theory,¹⁶ it is quite clear that it was a contention that was not by any means wholly accepted by the Congress that enacted § 15a (3). One of the specific examples given of an undesirable practice, and accepted by the members of the Commerce Committee that drafted the statute as such, was a case in which certain railroads had engaged in day-to-day differential pricing on the carriage of citrus fruit from Florida depending on whether competitive carriage was available by ship that day. See Hearings, supra, at 153-155. Similar complaints were made about seasonal variations in rates by railroads depending on whether winter conditions interfered with the carriage of freight by water. Id., at 162-163. Yet, from an economic standpoint, such rate variations make perfect competitive sense insofar as maximization of railroad revenues is concerned.¹⁷

The simple fact is that § 15a (3) was not enacted, as the railroads claim, to enable them to price their services in such a way as to obtain the maximum revenue therefrom. The very words of the statute speak of “preserving]” the inherent advantages of each mode of transportation. If all that was meant by the statute was to prevent wholly noncompensatory pricing by regulated carriers, language that was a good deal clearer could easily have been used. And, as we have shown above, at least one version of such clear language was proposed by the railroads and rejected by the Congress. If the theories advanced by the economists who testified in this case are as compelling as they seem to feel they are, Congress is the body to whom they should be addressed. The courts are ill-qualified indeed to make the kind of basic judgments about economic policy sought by the railroads here. And it would be particularly inappropriate for a court to award a carrier, on economic grounds, relief denied it by the legislature. Yet this is precisely what the District Court has done in this case.

We do not mean to suggest by the foregoing discussion that the Commission is similarly barred from making legislative judgments about matters of economic policy. It is precisely to permit such judgments that the task of regulating transportation rates has been entrusted to a specialized administrative agency rather than to courts of general jurisdiction. Of course, the Commission must operate within the limits set out by Congress in enacting the legislation it administers. But nothing we say here should be taken as expressing any view as to the extent that § 15a (3) constitutes a categorical command to the ICC to use fully distributed costs as the only measure of inherent advantage in intermodal rate controversies. As was stated in the New Haven case, it “may be” that after due consideration another method of costing will prove to be preferable in such situations as the present one. All we hold here is that the initial determination of that question is for the Commission.

It is in this connection that the timing of this case takes on particular significance. We have already observed that the ICC has presently pending before it a broad-scale examination of the whole question of the cost standards to be used where comparisons of inter-modal cost advantages are required. Rather than await the result of that rulemaking proceeding, the railroad appellees here determined to attempt to raise precisely the same issues in a much more circumscribed proceeding by unilaterally reducing their rates on one item of traffic. The District Court totally ignored the temporary nature of the ICC’s action in this case and the pendency of the rulemaking proceeding. Instead, it went ahead and, in the guise of resolving this particular controversy over a single rate reduction, rendered a decision which, for all practical purposes, made the rulemaking proceeding moot. While there might be some justification for such a course when the applicable statute clearly requires the agency to arrive at a given result, this case is emphatically not such a situation. As this Court stated in New Haven, “ [t] hese and other similar questions should be left for initial resolution to the Commission’s informed judgment.” 372 U. S., at 761.

The Commission stated here that it intended to exercise its informed judgment by considering the issues presented here in the context of a rulemaking proceeding where it could evaluate the alternatives on the basis of a consideration of the effects of a departure from a fully distributed cost standard on the transportation industry as a whole. Until that evaluation was completed, the Commission took the position that it would continue to follow the practice it had observed in the past of dealing with individual rate reductions on a fully distributed cost basis. The District Court, in effect, refused to permit the Commission to deal with the complex problems of developing a general standard of costing to use in determining inherent advantage in situations involving intermodal competition in the broad context of a rule-making proceeding. Instead, it ordered the Commission to resolve those problems in the narrow context of this individual rate reduction proceeding.

We have already observed that the District Court erred in interpreting the New Haven decision to require the Commission to permit out-of-pocket pricing in most instances. Given the fact that New Haven indicated that the Commission was to exercise its informed judgment in ultimately determining what method of costing was preferable, it is clear that the District Court also erred in refusing to permit the Commission to exercise that judgment in a proceeding it reasonably believed would provide the most adequate record for the resolution of the problems involved. We can see no justification for denying the Commission reasonable latitude to decide where it will resolve these complex issues, in addition to how it will resolve them. The action by the District Court here not only deprives the Commission of the opportunity to make the initial resolution of the issues but also prevents it from doing so in a more suitable context.

This Court has just recently held that the Federal Power Commission had the authority to fix rates on an area-wide basis rather than on an individual producer basis and that, in order to make such a procedure feasible, it had statutory authority to impose a moratorium upon rate increases by producers for a period of 2% years after the setting of the area rate. Permian Basin Area Rate Cases, 390 U. S. 747 (1968). The basis for this holding was the principle that the “legislative discretion implied in the rate making power necessarily extends to the entire legislative process, embracing the method used in reaching the legislative determination as well as that determination itself.” Id., at 776. That principle is equally applicable to rate regulation carried out by the ICC, especially where, as here, the determination made on an interim basis is in general accord with both the legislative history of the statute involved and the results in prior cases decided by the agency. Accordingly, we hold that the Commission had ample authority to decline to deal with the broad contentions advanced by the railroads in this individual rate case and that the District Court erred in failing to recognize that authority.

The District Court also objected to the failure of the Commission to explain why it permitted out-of-pocket ratemaking where the competing carrier was unregulated and not where the competitor was regulated. The short answer to this is that § 15a (3) by its own terms applies only to “modes of transportation subject to this Act,” which by definition means regulated carriers. As a result any arbitrariness that may flow from the distinction recognized by the Commission between regulated and unregulated carriers in situations of intermodal competition is the creation of Congress, not of the Commission.

The District Court also appears to have held that the Commission did not adequately explain how the rate set by the railroads would impair or destroy the barge-truck inherent advantage. Yet the Commission pointed out that the principle proposed by the railroads would, if recognized, permit the railroads to capture all the traffic here that is presently carried by the barge-truck combination because the railroads’ out-of-pocket costs were lower than those of the combined barge-truck service. The District Court seems to have been impressed by the fact that the railroads were merely meeting the barge-truck rate, despite the uncontroverted evidence that given equal rates all traffic would move by train. Given a service advantage, it seems somewhat unrealistic to suggest that rate parity does not result in undercutting the competitor that does not possess the service advantage. In any event, regardless of the label used, it seems self-evident that a carrier’s “inherent advantage” of being the low cost mode on a fully distributed cost basis is impaired when a competitor sets a rate that forces the carrier to lower its own rate below its fully distributed costs in order to retain the traffic. In addition, when a rate war would be likely to eventually result in pushing rates to a level at which the rates set would no longer provide a fair profit, the Commission has traditionally, and properly, taken the position that such a rate struggle should be prevented from commencing in the first place. Certainly there is no suggestion here that the rate charged by the barge-truck combination was excessive and in need of being driven down by competitive pressure. We conclude, therefore, that the Commission adequately articulated its reasons for determining that the railroads' rate would impair the inherent advantage enjoyed by the barge-truck service.

The judgment of the District Court is reversed and the cases are remanded to that court with directions to enter a judgment affirming the Commission’s order.

It is so ordered.

1 The United States, a statutory defendant in the District Court, supported the railroads’ position there and has participated in support of them in the proceedings before this Court.

2 The -term “inherent advantage” comes from the National Transportation Policy, 49 U. S. C. preceding § 1, and is incorporated by reference into § 15a (3) of the Interstate Commerce Act. The meaning of the term is the central issue in these cases and will be discussed in considerable detail, infra, at 579-594.

3 Fully distributed costs are defined broadly by the ICC as the “out-of-pocket costs plus a revenue-ton and revenue ton-mile distribution of the constant costs, including deficits, [that] indicate the revenue necessary to a fair return on the traffic, disregarding ability to pay.” New Automobiles in Interstate Commerce, 259 I. C. C. 475, 513 (1945).

4 The long-term out-of-pocket costs were computed under an ICC-sponsored formula which generally holds that 80% of rail operating expenses, rents and taxes are out-of-pocket in that they will vary with traffic. To this is added a return element of 4% on a portion of the investment (all the equipment and 50% of the road property), which is apportioned to all traffic on a proportional basis. Compare n. 3, supra.

5 This figure is not precisely a cost figure. Rather it is the barge fully distributed cost, plus the charge made for the truck portion of the service and the charge for barge-truck transfer. Since all parties seem willing to treat the figure as one of fully distributed cost for the barge-truck combination, no further mention will be made of its disparate elements.

6 Because the barge-truck rate of $5.11 was below the fully distributed cost of the service, Division 2 of the ICC initially concluded that the barge-truck combination had forfeited its right to claim that its inherent advantage of lower fully distributed cost was being impaired by the railroads’ setting of a matching rate. On reconsideration, the full Commission reversed this ruling by Division 2, observing that there was no evidence that the failure of the barge-truck rate to equal fully distributed cost was due to anything but the barge lines’ ignorance of the precise amount of their fully distributed cost for this service. This determination is not challenged here by any party and we express no opinion on it.

7 Out-of-pocket costs have been regarded generally in these cases as equivalent to what economists refer to as “incremental” or “marginal” costs. Accordingly we shall equate the terms likewise, although we have no intention of vouching for the accuracy of that equation as a matter of pure economics. Cf. n. 4, supra. Such costs are defined generally as the costs specifically incurred by the addition of each new unit of output and do not include any allocation to that unit of pre-existing overhead expenses.

8 A rate is compensatory in the sense used by the District Court any time it is greater than the out-of-pocket cost of the service for which the rate is set. The term fully compensatory is sometimes used to describe a rate in excess of fully distributed costs.

9 An illustration of such a case is the decision of the ICC that was reversed in the New Haven case. There the ICC had refused to permit the railroads to set a rate which was not only above their out-of-pocket cost for the service but was also above their fully distributed cost for approximately half of the movements involved. The Commission did not rely on a determination of which of the competing carriers had the inherent advantage as to costs, but instead decided broadly that the rate would eventually destroy the coastwise shipping industry and therefore should be prohibited. This Court held that, in general, the ICC was required to determine which of the competing carriers possessed the inherent advantage before a rate could be ordered cancelled in order to protect a carrier’s present rate. While the Court indicated that the Nation’s defense needs might permit protection of even a higher cost carrier in some cases, it held that the ICC had not adequately shown New Haven to be such a case.

10 The appellees also contend, and the District Court held, that the statements in the legislative history of § 15a (3) that Congress intended to compel the Commission to return to the approach to competitive rate regulation it had utilized in the case of New Automobiles in Interstate Commerce, 259 I. C. C. 475 (1945), indicate that out-of-pocket ratemaking was intended to be the rule in such cases. However, the passage quoted from New Automobiles simply states that the rates of one mode of transportation should not be held up merely to protect competing modes. It says nothing at all about inherent advantages.

The railroads argue that the basic thrust of the New Automobiles case was to compare costs on an out-of-pocket basis. And it is true that many of the comparisons there made were on that basis. However, an examination of what the Commission actually said and did in New Automobiles compels the conclusion that no flat rule of comparison of out-of-pocket costs was there laid down. For example, the Commission concluded that on the basis of a comparison with the railroads' out-of-pocket costs for shipping automobiles, the truckers were the lower cost mode only up to 120 miles. On a fully distributed cost comparison the truckers were the lower cost mode up to 230 miles. 259 I. C. C., at 528. After discussing at some length the concept of reasonable minimum rates, the Commission ultimately concluded that generally the truckers had the cost advantage at distances up to 200 miles and that the railroads should be permitted to set rates that would permit them to compete for the longer-haul trafflc. 259 I. C. C., at 539.

Given the fact that the Commission was dealing with an attempt by the truckers to get it to hold up railroad rates for distances even greater than 600 miles, it is not surprising that the issue of measuring inherent advantage as between fully distributed and out-of-pocket costs did not receive detailed consideration, since by either method the truckers were the low cost mode only up to a little more than 200 miles. Thus it cannot fairly be said that New Automobiles represents a considered choice between the two methods of cost comparison. Rather what it stands for is the principle emphasized in the New Haven case that the rates of one mode should not be held up to protect the revenues of a competitor without regard to which is the low cost carrier.

In any event, what matters so far as § 15a (3) is concerned is not what the Commission meant in New Automobiles but what Congress thought it meant in 1958 when the section was enacted. As we have shown in the text of the opinion above, Congress considered New Automobiles to stand for the principle that the rate structure of a competing mode should not be protected by the Commission simply to prevent it from losing business through competition.

11 The District Court also relied on the rejection of a similar proposal by truck and barge interests that fully distributed costs be the floor for reasonable minimum rates in the course of the enactment of the National Transportation Policy in 1940. It seems clear, however, that one of the major reasons for the rejection of the so-called Miller-Wadsworth amendment by Congress was the possibility that its enactment would prevent low-value industrial and agricultural commodities from being carried at a rate low enough to make it economically feasible to ship them in interstate commerce. See generally Nelson, Rate-Making In Transportation — Congressional Intent, 1960 Duke L. J. 221, 228-238.

12 While it is true that, for varying and sometimes unexplained reasons, the Commission has not invariably used fully distributed costs as the basis for cost comparisons in situations involving inter-modal competition, see 268 F. Supp., at 78, it is also true that it has generally declared fully distributed cost comparisons to be preferable. Thus in the hearings on the bill that was to become §15a(3), Commissioner Freas stated:

“Whenever conditions permit, given transportation should return the full cost of performing carrier service. ... In many instances, however, the full cost of the low-cost form of transportation exceeds the out-of-pocket cost of another. If, then, we are required to accept the rates of the high cost carrier merely because they exceed its out-of-pocket costs, we see no way of preserving the inherent advantages of the low cost carrier.” Quoted at 372 U. S., at 755.

13 The District Court ascertained the legislative purpose to promote “hard competition” from the following passage from the New Haven opinion:

“Section 15a (3), in other words, made it clear that something more than even hard competition must be shown before a particular rate can be deemed unfair or destructive. The principal purpose of the reference to the National Transportation Policy, as we have seen, was to prevent a carrier from setting a rate which would impair or destroy the inherent advantages of a competing carrier, for example, by setting a rate, below its own fully distributed costs, which would force a competitor with a cost advantage on particular transportation to establish an unprofitable rate in order to attract traffic.” 372 U. S., at 759.

Since the sentence following the term “hard competition” described an example of the competition prohibited by the National Transportation Policy that is identical to the facts of the present case, the District Court’s use of the term to reverse the ICC’s decision here seems somewhat peculiar.

14 Constant costs are, broadly speaking, those items of expense which are incurred by a business regardless of the scale of its operations. They are essentially the equivalent to what is commonly called overhead expenses. For railroads constant costs include such items as real estate taxes, certain rents, much right-of-way maintenance expense and similar expenses.

15 Unequal allocation of constant costs as an element of the rate charged also occurs commonly where a bulky commodity is so low valued on a per ton basis that setting a rate by reference to the fully distributed cost of carrying the commodity would make it uneconomic to ship it. See n. 11, supra.

16 This Court is not particularly suited to pass on the merits of the economic arguments made by the railroads’ expert witnesses in these cases. Moreover, their soundness is not especially relevant to the result we reach in the present posture of this controversy. However, because the economic testimony is emphasized so heavily by both the railroads and the United States in their arguments to us, we shall venture a few observations on it.

Most of the economic testimony is directed towards proving that the utilization of out-of-pocket costs in setting rates permits the railroads to maximize their profits. To the extent that out-of-pocket costs are accurately computed, that proposition appears uncontro-vertible. The economists then go on to argue, in effect, that what is good for the railroads is good for the country. This argument is developed as follows. Whenever a railroad lowers its rate, the shipper to whom that rate is available benefits. As long as the rate is above the out-of-pocket cost of the service, the railroad benefits by obtaining the profits from traffic it formerly did not carry. The fact that a competing carrier may lose the revenue it previously earned by carrying the traffic is immaterial because the railroad’s ability to make a profit by charging the lower rate shows that it is, in some sense, more efficient than its competitor.

In order to evaluate the foregoing argument certain other aspects of a railroad’s operation must be kept in mind. The reason why a railroad’s fully distributed costs are substantially greater than its out-of-pocket costs on any given traffic is, inter alia, because certain constant costs, see n. 14, supra, are allocated to that traffic on a proportional basis despite the fact that those costs will be incurred by the railroad whether it carries the particular traffic or not. These constant costs must be earned if the railroad is to stay in business. They are allocated proportionally on the theory that, all other factors being equal, such an allocation will be the best way of assuring that each shipper contributes his fair share towards covering the constant costs. Obviously to the extent that any shipper pays more of the constant costs than another without any good reason for so doing that shipper is, in some sense, discriminated against.

The railroad economists point out that, because constant costs by definition are not attributable to the carriage of any particular traffic, it is to some extent arbitrary to allocate them to particular traffic. They further contend that all shippers presently utilizing a railroad’s services are benefited when the railroad obtains additional traffic at a profit to it, because that profit can be used to pay a portion of the constant costs currently being charged wholly to them. The fact that charging a rate less than its fully distributed cost of carrying the traffic results in the shipper of that freight paying a disproportionally low share of the railroad’s constant costs is considered to be outweighed by the overall benefit to the other shippers of having the absolute amount borne by them of the constant costs decreased by the profit earned on the traffic.

It seems apparent, however, that in a case where the sole reason that a rate below fully distributed cost is necessary to attract such additional traffic is the competition of another mode of transportation, the continued existence of that competition is also the sole economic justification for maintaining the rate at a relative level that favors one shipper over others. If the competing carrier is driven out of business because of its inability to match the railroad’s lower rates set on an out-of-pocket basis, the economic justification for permitting the continuation of those low rates would seem to disappear. Yet the railroad economists assert that in such a situation the railroad should be required by the ICC to maintain the rate at its original level. The obvious reason for this position is that permitting a railroad to raise its rates once it had effectively destroyed a competitor in one area would enable it to price on an out-of-pocket basis in competition with another carrier in a different area thereafter and, in turn, drive that carrier out of business. Eventually a railroad could eliminate all its competitors whose out-of-pocket costs were higher than its own. After this was accomplished the railroad could re-price all its services on a fully distributed cost basis thereby eliminating all discrimination between its shipper customers.

Of course, the shippers formerly served by competing modes at rates profitable to them but lower than the railroad’s fully distributed costs would at that point have lost the advantage of the low cost service. The only way to perpetuate the advantage previously enjoyed by those shippers would be, as the railroad economists recognized, artificially to maintain their rates at the former level despite the absence of present economic justification for such a low rate. (It is true that were the barriers to re-entry into the transportation market low, as asserted by the railroad economists, the potential competition created by the possibility of such re-entry by a competing mode could furnish an economic justification for the continuance of the original low rate. However, there is no factual evidence in this record from which it can be concluded that barriers to re-entry are low enough to create such potential competition.)

If the only justification for the maintenance of a disproportionally low rate to some shippers is the fact that competition existed once upon a time for their business, would it be irrational to conclude that it would be preferable to keep the original competition in business to serve those shippers and to require the railroad to look elsewhere for additional revenue? Would it, for example, be possible for railroads to increase their revenues instead by increasing, through selective rate decreases, the volume of traffic shipped by persons who presently pay amounts in-excess of the fully distributed cost for the service afforded them? These are only a few of the questions that come to mind when we attempt to evaluate the economic arguments made in this ease. We do not pretend to be able to answer them. We merely note their existence as evidence that we do not find the arguments made to the ICC here as compelling as did the District Court.

Our discussion here should not be interpreted as a rejection of the basic economic points made by the railroads. It is merely intended to illustrate the desirability of having the initial resolution of these issues made by a tribunal, and in a proceeding, more suitable than the present one.

17 It is, of course, true that such discriminations need have no necessary relationship to a railroad’s cost of service, whether that is computed on a fully distributed or out-of-pocket basis. On the other hand, it is also evident that what is basically at issue is a carrier’s right to price discriminatorily, either between shipments or shippers, in order to maximize revenues by competition. By contrast it can be noted that the railroads have apparently retained their prior rate of $11.86 per ton on ingot molds in areas where they have no competition from barge-truck service. The discrimination thus created is not too dissimilar from that embodied in the above examples.