New York, New Haven & Hartford R. Co. v. United States, 199 F. Supp. 635 (D. Conn. 1961)
November 15, 1961
UNITED STATES of America and Interstate Commerce Commission, Defendants,
Sea-Land Service, Inc., and Seatrain Lines, Inc., Defendants-Intervenors.
United States District Court D. Connecticut.
Carl Helmetag, Jr., Philadelphia, Pa. (Thomas P. Hackett, and Eugene E. Hunt, New Haven, Conn., Andrew C. Armstrong, Baltimore, Md., Edwin N. Bell, Houston, Tex., James A. Bistline, Washington, D. C., Ernest D. Grinnell, Jr., St. Louis, Mo., J. Edgar McDonald, *637 New York City, Clarence Raymond, Louisville, Ky., Charles P. Reynolds, Richmond, Va., Albert B. Russ, Jr., Jacksonville, Fla., and Toll R. Ware, St. Louis, Mo., of counsel), for plaintiffs.
B. Franklin Taylor, Jr., Associate Gen. Counsel, I. C. C., Washington, D. C., Richard H. Stern, Atty., Dept. of Justice, Washington D. C. (Lee Loevinger, Asst. Atty. Gen., Dept. of Justice, Washington, D. C., Harry W. Hultgren, Jr., U. S. Atty., D. Conn., Hartford, Conn., and Robert W. Ginnane, Gen. Counsel, I. C. C., Washington, D. C., of counsel), for defendants.
Warren Price, Jr., Washington, D. C. (Albert W. Cretella, New Haven, Conn., and William H. Ambrecht, Jr., Mobile, Ala., of counsel), for defendant-intervenor Sea-Land Service, Inc.
Ralph D. Ray, New York City (Morris Tyler, Gumbart, Corbin, Tyler & Cooper, New Haven, Conn., Chadbourne, Parke, Whiteside & Wolff, Alan S. Kuller, and Charles J. Prentiss, New York City, of counsel), for defendant-intervenor Seatrain Lines, Inc.
Before HINCKS, Circuit Judge, and ANDERSON and TIMBERS, District Judges.
HINCKS, Circuit Judge.
This is an action brought by several railroads to enjoin an order of the Interstate Commerce Commission, entered pursuant to a report published in 313 I.C.C. 23, directing them to cancel substantial rate reductions for 66 listed movements of their trailer-on-flat-car (TOFC) service between points in the East and Texas. Two water carriers, Sea-Land Service, Inc. (hereinafter referred to as Sea-Land), and Seatrain Lines, Inc. (hereinafter Seatrain), protested the proposed rates and appeared herein to defend the order below. The United States and the Interstate Commerce Commission also appeared herein and defended the order. They will be referred to collectively as "the government."
Sea-Land, known in the proceedings before the Commission by its former name Pan-Atlantic Steamship Corporation, in 1957 had suspended its domestic break-bulk freight service, theretofore operated between eastern ports and Southern Atlantic and Gulf ports. It substituted four "trailer-ships," each of a capacity to carry 226 standard, demountable, truck trailer-bodies, and each equipped with cranes capable of lifting the loaded trailers from their chassis on the pier, stowing them aboard, unopened, into cooperating slots and, at the ports of destination, lifting them onto trailer-chassis on the pier. By this improved highway-water-highway "fishy-back" service, Sea-Land offered a door-to-door service to all shippers and consignees accessible by highway in containers locked or unopened between the point of origin and destination. The conversion from break-bulk service to the "fishy-back" service just described enabled Sea-Land to improve the quality of its service and reduce operating costs at rates which, prior to 1957, were five to ten per cent below the rail boxcar rates.
Seatrain offers rail-water-rail service whereby loaded railroad cars are taken aboard its three steamships at Edgewater, New Jersey and after carriage by sea to a destination in a southern coast or a Gulf port are then carried by a railroad for delivery to the consignee. Seatrain contemplates a modification of this service, a so-called "seamobile service," whereby the freight will be carried in special containers which may be readily transferred from highway trailers or railcars to and from its seagoing vessels. Like railroad boxcar service, the present Seatrain service permits carriage from shipper to consignee without breaking bulk only when shipper and consignee are located on railroad sidings.
*638 To compete with these services and especially that of Sea-Land which is available to shippers and consignees without rail access, the railroads considerably extended their "piggy-back," highway-rail-highway, service, whereby trailer-bodies, without detachment from their chassis, are hauled onto and tied down upon railroad flatcars, one or two to each flatcar, and at the rail destination are hauled by tractors to the consignees' doors. Prior to 1957 the rates set for this TOFC service were generally on a parity with those in force for regulated motor-carrier service and somewhat higher than the rail boxcar rates. But to make their competition with Sea-Land's fishy-back service more successful, the railroads in 1957 filed rate schedules for their TOFC service which were substantially on a parity with Sea-Land and Seatrain rates, R. 33, and motor common carrier rates, R. 45. However, these rate schedules, since inaugurated as an experiment, were limited to 66 commodity movements from particular eastern points to Fort Worth and Dallas and return.
On petitions of Sea-Land, Seatrain, a motor-carrier association, the Secretary of Agriculture, and several municipal authorities, the Commission placed all the TOFC rates in this initial, or pilot, schedule under suspension and investigation under I. & S. Docket No. 6834 "Piggy-back RatesBetween East and Texas," which also covered the lawfulness of Sea-Land rates between the same points and certain Seatrain rates. This controversy, together with three others involving the lawfulness of numerous other Sea-Land rates, each under separate docket numbers, I. & S. Docket No. M-10415, I. & S. Docket No. 6906, and I. & S. Docket No. M-11375, came on for hearing before Examiner Morgan who filed a separate report on each. On exceptions by the parties, No. M-10415 was heard by Division 3 of the Commission and on further exceptions by the railroads to the Division 3 report all four docket numbers were consolidated for hearing and dealt with in a consolidated report by the entire Commission.
The Commission held that the entire schedule of TOFC rates was unlawful and, in the order under attack herein, directed that the rates, which had theretofore been under suspension, be canceled. The cancellation date, however, was ordered suspended, thus leaving the rates in continuing suspension. It was to set aside the cancellation order as to the TOFC schedule that the railroads brought this action. Except for a few specific rates, the Sea-Land and Seatrain rates were found lawful and to this holding no exception had been taken to the Division 3 report.
The Commission's essential findings as enunciated by five of the Commissioners in its report were as follows:
1. The proposed TOFC rates would produce revenues exceeding out-of-pocket costs (see Appendix A, infra) for all of the proposed movements by TTX flatcars and for all but six of 66 of the listed movements by railroad-owned cars, and exceeding the railroads' fully-distributed costs (see Appendix A) for *639 43 of the 66 movements by TTX cars and 14 movements by railroad-owned cars. R. 36. Consequently, except for the six rates returning less than out-of-pocket costs the TOFC rates were found to be compensatory. The corresponding Sea-Land rates, with one exception, were similarly found to be compensatory, as were the Seatrain rates.
2. As to the 66 movements in issue here, Sea-Land costs, both on an out-of-pocket and a fully-distributed basis, were lower than TOFC costs except for two (out of the 66) movements accomplished by TTX cars. However, railroad boxcar costs on some of this traffic were lower than Sea-Land costs; on other portions of the traffic, Sea-Land costs were lower. On the record before it the Commission said it "can not determine * * * where the inherent advantages may lie as to any of the rates in issue. R. 46.
3. In quality of service, the several modes rated in the following order: TOFC, all-rail boxcar, Sea-Land and Seatrain. However, "the most important, and usually the determinative, factor to the shippers as a whole is the measure of the rates." R. 29.
4. All Sea-Land traffic is competitive with the railroads. But only a fraction of railroad traffic is competitive with Sea-Land. R. 45.
5. The railroads intend, if the proposed TOFC rates attract profitable traffic, to extend the reductions in TOFC rates to many other movements than the 66 immediately involved herein. R. 47.
6. Sea-Land must recover fully-distributed costs to remain in business. R. 38.
On conclusions thought to follow from these findings, the Commission ordered the proposed TOFC rates cancelled, holding that "the rail TOFC rates on the commodities from and to the points concerned in I. & S. No. 6834 should be maintained on a level no lower than 6 percent above * * * sea-land rates, so long as the latter are not increased above their present levels." And its order was expressly stated to be "without prejudice to the filing of new schedules in conformity with the conclusions herein." R. 50. (Emphasis supplied.)
We hold that, at least on this record, the requirement of a rate differential to protect the water carriers violated the 1958 Amendment to the Interstate Commerce Act, now appearing as 49 U.S.C.A. § 15a(3), which reads as follows:
"In a proceeding involving competition between carriers of different modes of transportation subject to this Act, the Commission, in determining whether a rate is lower than a reasonable minimum rate, shall consider the facts and circumstances attending the movement of the traffic by the carrier or carriers to which the rate is applicable. Rates 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."
In disapproving the proposed schedule of the railroads the Commission, contrary to the specific prohibition of the 1958 amendment, is plainly holding up railroad rates "to protect the traffic" of *640 another mode. It argues, however, that the national transportation policy (hereinafter sometimes refered to as N TP), to which it is commanded to give "due consideration" by the same provision, compels this result. The evidence and findings do not support this argument.
The first policy-factor mentioned in the NTP declaration is to "recognize and preserve the inherent advantages of each [mode of transportation]." By "the inherent advantages" was meant the ability of a mode of transportation over the long run to provide a transportation service more acceptable to its shippers, by reason of quality or price, than that offered by a competing mode. That the calculation was to be long-run must be emphasized. The shorter-run "out-of-pocket" costs of one mode (e. g., railroads) may be lower than the longer-run "fully-distributed," or even the shorter-run costs of competing modes (e. g., water carriers) whose long-run costs are lower. When they are, rates set by reference to out-of-pocket costs may favor what in the long run is the less efficient, higher-cost mode. Thus the "inherent advantages" of lower cost (or better service, which is discounted for price) refers to the long-run, or fully-distributed, costs of carriage. It was thought undesirable that one mode should undercut the rates of a competing lower-cost mode. Such conduct savored *641 of a predatory competitive practice. See Dixie Carriers v. United States, 351 U.S. 56, 59, and n. 5, 76 S. Ct. 578, 100 L. Ed. 934 (1956). And generally it was possible to destroy a lower-cost carrier or mode only by reducing rates, at least temporarily, to a level below the costs of either. It well may be that for the higher-cost mode to jeopardize the continued existence of a lower-cost mode by setting rates below the costs of each, could be characterized as a "destructive competitive practice" which under another NTP policy-factor was not to infect the "reasonable charges for transportation services" which the Commission was authorized to approve. But that is not this case, as we will now proceed to show.
The Commission in its report expressly admitted (R. 46) that "we can not determine on these records where the inherent advantages may lie as to any of the rates in issue." It thus made it plain that it did not rest its holding on the "inherent advantages" factor of the NTP. Instead, it turned to another NTP factor to justify its decision. It said, at R. 44, the "TOFC rates here under investigation, with the few exceptions noted, appear to be compensatory. Many of these are above the fully-distributed costs shown, and with the exceptions mentioned, all are above out-of-pocket costs. The next, and the most important, question is whether these rates constitute destructive competition." (Emphasis supplied.) And that the Commission did, at least in part, base its decision upon a holding that the proposed TOFC rates were an "unfair or destructive competitive practice" such as it thought frowned upon by the NTP, is demonstrated by its statement that: "The reduced rates of the railroads here under consideration are an initial step in an overall program of rate reductions that can fairly be said to threaten the continued operation, and thus the continued existence, of the coast-wise water-carrier industry generally," R. 47. Apparently the Commission thought that any rate-competition which threatens the continued existence of a competitor it had power to prevent as a "destructive competitive practice," irrespective of whether the challenged rates were compensatory to the proponent thereof or whether the mode of the contesting competitor was a lower-cost mode than that of the proponent. This, we hold, was an erroneous interpretation of the Act, as amended.
Even before 1958, it would have been erroneous to hold that irrespective of other factors rates were unlawful merely because they would destroy carriers operating under different modes of transportation. In Schaffer Trans. Co. v. United States, 355 U.S. 83, 91, 78 S. Ct. 173, 178, 2 L. Ed. 2d 117 (1957), it was said that "[t]he ability of one mode of transportation to operate with a rate lower than competing types of transportation is precisely the sort of `inherent advantage' that the congressional policy requires the Commission to recognize."
Prior to 1958, the Commission's emphasis as between different factors of the NTP had vacillated; now allowing rate reductions to "recognize and preserve * * * inherent advantages," see New Automobiles in Interstate Commerce, 259 I.C.C. 475, and later shifting its stress to "coordinating * * * a national transportation system" even at the cost of stifling competition, as in A. W. Schaffer ExtensionGranite, 63 M.C.C. *642 247, rev'd sub nom. Schaffer Trans. Co. v. United States, supra.
Against this background of vacillation, the Transportation Act of 1958 was adopted. We think that, on the whole, the amendment of Section 15a was intended to provide freer play for competition as between different modes while still continuing protection to the modes having the inherent advantage of low cost from unfair or destructive competitive practices. In this the Congressional history, while perhaps not conclusive, bears us out. In H.R. Rep. No. 1922, 85th Cong., 2d Sess. (1958), 2 U.S.Code Congressional and Administrative News, p. 3456, 3470 (1958), it had been said: "* * * The Interstate Commerce Commission has not been consistent in the past in allowing one or another of the several modes of transportation to assert their inherent advantages in the making of rates." This House Report on § 15a(3) in its final form said, further: "The effect of this amendment will be to encourage competition between the different modes of transportation for the benefit of the shipping public." It quoted with approval the above excerpt from the opinion in Schaffer Trans. Co. v. United States, supra.
The report of the Senate Committee of June 3, 1958 (S.Rep. No. 1647, 85th Cong., 2d Sess.) said of § 15a(3) in its final form,
"The committee wishes further to emphasize that the amendment in regard to section 5 amending section 15a of the act as framed by the committee is designed to encourage competition in transportation by allowing each form of transportation subject to the Interstate Commerce Act full opportunity to make rates reflecting the inherent advantages each has to offer, with such rate-making being regulated by the Interstate Commerce Commission, however, to prevent `unfair or destructive competitive practices' as contemplated by the declaration of national transportation policy. Under the committee amendment the principal emphasis, but not the exclusive emphasis, in a competitive ratemaking proceeding involving different modes of transportation will be on the conditions surrounding the movement of the traffic by the mode to which the rate applies." (Emphasis supplied.)
Having in mind that by the 1958 amendment Congress for the first time articulated an express, though qualified, prohibition against holding up the rates of one mode to protect another, a prohibition accompanied by a direction that consideration shall be given to the effect of rates on the carrier for whom they are prescribed, we are unable to accept the contention in the government's brief that "section 15(a) (3) brought about no fundamental change in the law." We think that Congress in adopting the 1958 amendment, which its committees thought would encourage competition, did not intend the included prohibition of compulsory rate differentials to be nullified merely because of the adverse effect of rate competition on another mode of transportation even if carried to the point of rendering one mode of transportation obsolete and hence unable to survive. Instead, we think, the differential prohibition was intended to be qualified only when factors other than the normal incidents of fair competition intervened, such as a practice which would destroy a competing mode of transportation by setting rates so low as to be hurtful to the proponent as well as his competitor or so low as to deprive the competitor of the "inherent advantage" of being the low-cost carrier. The "inherent advantage" factor and the "destructive competitive practice" factor were the only two policy factors mentioned in the committee reports. We think the reference to the NTP in § 15a (3) indicates an intent that the differential prohibition was to be qualified only when necessary because of the interplay of these two factors.
We cannot help wondering if the Commission's obvious reluctance to accept as *643 critical the relative costs of service by the competing modes of transportation may not be attributable less to its interpretation of the applicable law than to its fear that the process which it has developed of so-called value-of-service ratemaking will be jeopardized if it be required to make critical findings as to the comparative costs of competing modes of transportation.
Value-of-service ratemaking is a price-discrimination device, used either to maximize profit or to subsidize certain interests. As a maximizing tool, it can be used by a monopolist in the following way. M has a producttransportation with two potential buyers, A and B. A uses it to ship industrial sand which he sells for $10 a ton, B to ship coal which he sells for $35 a ton; M's cost of shipment is the same for both. If he sets a uniform rate per ton of $2, B will ship but A will nothe will not be able to meet his competition at the market. If M sets his price at a uniform $1 per ton, both will shipbut M will lose a greater revenue which he could have garnered from B. And if M's out-of-pocket cost of carriage is $0.75 per ton, he will want A's $1 traffic. Early railroads thus developed the practice of charging $2 to B and $1 to A, cost of carriage notwithstanding.
This value-of-service pricing is not necessarily unsound economically. Economists generally agree that:
"Preferential rates relieve rather than burden other traffic if two conditions are fulfilled. These are (1) that the rate must more than cover the direct costs; and (2) that the traffic will not move at higher rates."
And the I.C.C., partly because this was the rate pattern prevailing when the Commission was established, and partly to maximize utilization of the railroads, adopted value-of-service as its own criterion of ratemaking.
The Commission, however, added factors of discrimination other than profit maximization. One of these is subsidization of certain commodities producers, perhaps under political pressures. Thus corn is carried at 85 per cent of out-of-pocket costs; beets at 57 per cent; gravel and sand at 88 per cent; logs at 62 per cent. And of course, passenger traffic has been carried at a loss for some time. These losses are made up on other traffic, such as gasoline, 135 per cent; equipment parts, 236 per cent; and metal alloys, 264 per cent. Other discriminations sometimes introduced by the I.C.C. are attributable to its desire to act as an economic planner, see, e. g., Anchor Coal Co. v. United States, 25 F.2d 462, 470 (S.D.W.Va.1928).
These official discriminations, hallowed and encrusted by time and inertia, now pervade the rate structure; indeed they are the rate structure. The rates on high-value commodities are thus twice subject to arbitrary boosting. First, they bear as part of their "costs" the subsidies extended to traffic which does not pay the out-of-pocket cost of its carriage. And second, high-value commodities are expected to yield for the carriers a profit above their fully-distributed costs sufficient to compensate for any deficiency in the fully-distributed costs of other traffic.
This disquisition may help to an understanding of what the I.C.C. attempted and did in this controversy. It serves to point out that "cost" as used by the I.C.C. is a term of art. "Fully-distributed costs" of operation are actually the I.C.C.'s estimate of returns necessary to continued profitable operation. "Fully-distributed cost" of a given service is a largely arbitrary estimate of the proportion of system "costs" which the I. C.C. feels a given service should contribute *644 to total revenue. For example, under the Commission's scheme of ratemaking the "fully-distributed cost" of the TOFC freight service is set at a level high enough to absorb, in addition to direct freight expenditures, the TOFC "share" of the eastern railroads' passenger-operations deficit. See Appendix A.
Thus the Commission was inhibited, by the terms of its own analysis, from a meaningful comparison of TOFC's cost to Sea-Land's cost. The "costs" of TOFC for these 66 movements were the sum of the following components: direct TOFC costs, a judgment of the appropriate shares of subsidy owed by TOFC to other railroad traffic, and an estimate of what contribution to total railroad revenue the high-value commodities here involved should make. Except for the application of value-of-service criteria Sea-Land "costs" were not complicated by these extrinsic factors. In short, under the Commission's scheme, TOFC costs and Sea-Land costs were incommensurate quantities.
But this lack of real significance in its cost comparisons is almost academic when set against the Commission's method of decision. That process, as nearly as we can trace it, was as follows. First, Sea-Land's overall revenue needs were calculated by the process outlined in Appendix A. Value-of-service criteria were then applied to determine the share of its needs to be met by the commodities in question. From this figure was derived the "fully-distributed cost" of Sea-Land's carriage of these items. The railroads, in effect, were then ordered to leave the traffic with Sea-Land by maintaining their rates at a 6 per cent differential. This conclusion was bolstered by a recital of TOFC "costs" computed as we have seen on an essentially different basis. The effect was to obscure, rather than illuminate, the identification of the "low-cost" mode.
If, instead of cancellation, the disposition of the proposed rates had been one of approval, and forced Sea-Land to reduce its own rates on the commodities involved, it is true that the substantial extra margin of profit the Sea-Land rates provide for the company as a whole, which is available to augment the lower profits from the revenues of low-value movements, would have been reduced. This, we think, is what the Commission meant when it said: "The reduced rates of the railroads here under consideration are an initial step in an overall program of rate reductions that can fairly be said to threaten the continued operation, and thus the continued existence, of the coastwise water-carrier industry generally." R. 47. Apparently, the Commission thought that any competition having such effect on water carriers, i. e., to reduce rates on high-value commodities, was necessarily "an unfair or destructive competitive practice" within the meaning of the statutory declaration of the NTP.
Such an interpretation of the Commission's decision might avoid some of the difficulties we find with that order; for example, it would explain why the Commission refused to allow TOFC rates to be lowered to a level still above the railroads' fully-distributed costs. But if the Commission decision does represent such an attempt to preserve its value-of-service ratemaking structure, the decision must fall for lack of evidence and necessary findings. For the Commission would have been warranted in holding up the TOFC rates for these 66 movements to protect an integral overall rate structure for Sea-Land only on evidence and findings that, notwithstanding the § 15a(3) prohibition of differentials, the integral overall rate structure was required by one or more of the policy-factors enumerated in the NTP, and particularly the policy to protect the "inherent advantages" of the overall structure against "destructive competitive practices" and "the need, in the public interest, of adequate and efficient railway transportation service at the lowest cost consistent with the furnishing of such service," § 15a(2).
There was no finding here that the overall rate structure of Sea-Land *645 which the Commission sought to preserve was that of the overall low-cost mode. This finding the Commission refused to make; indeed it could not, for as it said, "we do not have before us the rail costs as to many of these rates * * * we can not determine on these records where the inherent advantages may lie as to any of the rates in issue." True, it found that as to many of the 66 movements in question, Sea-Land was a lower-cost mode than TOFC; but it did not find that Sea-Land service was in general a lower-cost mode than railroad service in general. All-rail boxcar service, as well as TOFC, competes with Sea-Land, and as to many boxcar rates the Commission's finding was "the railroads' costs appear to be lower." R. 46. Yet the Commission concluded that Sea-Land was entitled to protection against boxcar competition, as well as TOFC competition, though by a differential "somewhat lower" than the 6% prescribed for TOFC. R. 50. See also I.C.C., I. & S. Docket No. 7454 (May 18, 1961). We conclude that, for lack of evidence and findings that Sea-Land was in general the low-cost mode, the action of the Commission, in so far as it sought to protect Sea-Land's overall rate structure, was an unlawful interference with the forces of competition fosterednot proscribedby § 15a(3).
For its disregard of the differential prohibition the Commission also places reliance on the "national defense" clause of the NTP. In this too, we think the Commission misinterpreted the law. True, the hoped-for "end" of the National Transportation Policy is a system "adequate to meet the needs * * * of the national defense." But the national defense is not stated as an operative policy or means: it is mentioned only as the hoped-for "end" of the operative policy-factors previously enumerated. It is not stated, and we think not intended, as a blanket grant of power to the Commission effective to nullify the express prohibition of rate differentials. By its emphasis on the supposed needs of national defense the Commission overrides the express prohibition of rate differentials without invoking, and without basis for invoking, the two policy-factors of the NTP, discussed above, which may properly qualify that prohibition. Its stress on national defense also brings it into conflict with all three policy-factors enumerated in § 15a(2), viz., the effect of the TOFC proposed rates on the TOFC carriers, the public need of railroad service at the lowest compensatory rate, and the carriers' need for compensatory revenues. These three factors are as much a part of the national policy as the several policy-factors in the NTP declaration. Even if these factors were deemed to conflict with each other, we think it not proper to disregard the more recent expressions of Congressional intent contained in paragraph (3) of § 15a.
Moreover, we find scant basis of fact supporting the Commission's action even if its interpretation of the applicability of the "national defense" clause were correct. The pertinent evidence seems to be confined to: (1) a 1955 report of the U. S. Maritime Administration entitled "Review of the Coastwise and Intercoastal Shipping Trades." This report stresses the need for break-bulk capacity. Neither Seatrain nor Sea-Land now have such capacity. (2) S.Rep. 2494, 81st Cong., 2d Sess. (1950), which refers in passing to the importance of coastal shipping to the national defense. But the Senate Report is eleven years old, and does not seem to have resulted in legislation. Whatever its relevance in 1950, it must yield to the specifics of the 1958 Act. After all, so far as appears, compulsory rate differentials were not set up to protect the coastwise carriers at the expense of competing modes and the shipping public, from their tremendous loss of tonnage between 1940 and 1958. R. 27.
The Commission says that "shipper evidence * * * is indicative of a need by the general public for the services of these lines." R. 49. This statement assumes the matter for decision: since cost is a major factor, as long as the Commission maintains a rate differential *646 shippers will "need" the lower-rate mode. Should the Commission heed the statute and allow the reductions, the "need" in evidence may well disappear.
The defendants' other arguments can be disposed of shortly.
The Commission indicates, and the government insists, that a full-fledged rate war is the inevitable consequence of allowing the proposed rate reduction. But surely, under its power to fix minimum rates, 49 U.S.C.A. § 15(1), the Commission will have power to disapprove rates not compensatory. Nothing in this opinion disparages that power.
Seatrain in its brief expresses fear that the proposed rates, if effective, would eliminate it as a competitor and suggests that the railroads might use its demise as an opportunity for rate increases. That bogie is laid at rest by 49 U.S.C.A. § 4(2). In other respects, the Seatrain position is largely that of Sea-Land and the government except that Seatrain produced no evidence from which its costs could be found. R. 38.
The government further argues that the proposed rates were disallowed not because a differential was needed to protect the water carriers but because the railroads failed to sustain the burden of proof. Thus it says in its brief: "when the proponent of a lowered rate which will deprive another mode of needed traffic, indeed deprive it of any opportunity to compete, fails to establish that the proposed rates are not cost-justified in that they reflect inherent cost advantages, then the rate should not, and certainly need not, be allowed." Beyond doubt, in the situation here, the burden was upon the railroads, as proponents, to submit evidence from which their own costs for the 66 movements could be determined. Indeed, the Act, 49 U.S.C.A. § 15(7), provides that "the burden of proof shall be upon the carrier [who seeks a rate-change] to show that the proposed changed rate * * * is just and reasonable." But where, as here, a rate-change is protested because of its effect upon a competing mode by one who claims to have the inherent advantage of lower costs, it is for the protestant to show its costs. This was expressly recognized in the dissenting report herein and not disputed in the majority report. And in the report of the Commission in No. 32920 on "Various Commodities from or to Arkansas and Texas," decided June 22, 1961, this rule was expressly recognized.
Finally, the government urges that even if the Commission's order was improperly based on a belief that the water carrier traffic, notwithstanding § 15a(3), should be protected, it would be futile and fruitless for us to set aside the order because the same order would then be made upon another ground, viz., that the water carriers were in fact the low-cost mode. As we have shown, no adequate basis for such a ground of decision is disclosed in the present record. But even if in this we are wrong, we cannot say what disposition the Commission would make of the controversy if authoritatively advised that the rationale of the report now before us is erroneous. In that event, it would be for the Commission, not this court, to decide whether the record should be reopened for further evidence and to evaluate the evidence. Securities and Exchange Commission v. Chenery Corp., 318 U.S. 80, at pp. 87, 88, 63 S. Ct. 454, 87 L. Ed. 626 (1943).
Accordingly, the order requiring cancellation of the TOFC rates is set aside, and the Commission is enjoined from cancellation of TOFC rates which return at least the fully-distributed cost of carriage.
If, however, on some enlarged record the Commission shall find that the water carriers are in general the low-cost mode, and if it shall also find that value-of-service considerations demand water carrier rates on particular movements and commodities which each return to the water carriers more than their fully-distributed costs, our injunction will not go so far as to prevent the Commission from requiring that TOFC *647 rates be set high enough to protect water carrier traffic; provided, however, a railroad rate for a particular movement, if it yields the railroad's fully-distributed cost which is lower than the water carrier's fully-distributed cost, may not be disturbed. It is noted that at least two of the rates involved in this proceeding were found to fall in this two-pronged category. See R. 35.
A decree in accordance with this opinion may be submitted by the plaintiffs, on notice unless consultation amongst the parties shall bring about a waiver of notice.
The I.C.C. report speaks freely of "out-of-pocket" costs and of "fully-distributed" costs but gives no definition to the sense in which it uses these terms. Since an understanding of the I.C.C.'s cost techniques is necessary to an understanding of its decision, we state in this appendix our understanding of the I.C.C. cost technique as gleaned from the sources indicated.
The I.C.C. employs different techniques for estimating railroad and water-carrier costs, reflecting the differing characteristics of each mode.
Water-carrier costs are calculated by traditional accounting methods, since they are "largely associated with the individual accounting unit," Meyer, Peck, Stenason & Zwick, Competition in the Transportation Industries, 112 (1959) i. e., the operations of a given ship or ships. The costs for component items including steamships, trucks, stevedoring charges, port charges, interest and depreciation, the motor service charges at each end of the voyage, and overhead are calculated for representative voyages. Costs per ton-mile are computed by applying average figures for tonnage and mileage.
Costs are then broken down into two categories, (1) "out-of-pocket" and (2) "fully-distributed." "Out-of-pocket" costs represent a rough approximation of the long-run marginal (i. e., added) costs of carriage. Estimates are made of the degree to which each of the cost components mentioned abovesteamships, trucks, stevedoring, etc.varies with the volume of traffic carried. To arrive at a ton-mile figure, non-varying expenditures (those attributable to size of plant rather than intensity of use) such as billing, terminal supervision, garages, etc., are subtracted from total expenditures, and the remainder divided by ton-miles carried. See Rationale of Cost Finding Section, Hearing Examiner's Proposed Report, I. & S. Docket No. M-10415, Appendix B.
"Fully-distributed" costs, by contrast, represent an approximation of the total long-run costs of remaining in operation. To obtain fully-distributed cost, corporate overhead is first added to "full" operation cost, i. e., to all direct expenditures. Then, on the basis of a 95% "operating ratio," i. e., the ratio between direct expenditures plus overhead, and fully-distributed costs, an allowance for profit is calculated which amounts to about 5.3% (.0526+) of direct expenditures plus overhead. The fully-distributed cost is the sum of this profit, the overhead, and direct expenditures. This method is used by the I.C.C. whenever capital investment is low, so that a return figure based on such investment would not reflect the size of operations. Sea-Land pressed this form of calculation because its vessels are largely chartered and its motor service hired. In this, the I.C.C. acquiesced. See Rationale of Cost Finding Section, I. & S. Docket No. M-10415, supra.
Railroad cost figures cannot be developed so easily. The basic difficulty is in "multiple use"the tracks, rolling stock, terminal expenses, and even the trains themselves, carry mixed shipments to different destinations. Apportionment of costs to the different services offered is a complex and difficult process. For example, assume a train already made up and ready to go. What is the marginal or added cost of tacking on two more flatcars? Or should these two cars bear some proportion of costs incurred before they were added, which *648 indeed would have been fully incurred even if the cars had not been added?For accounting purposes, the cars are regarded as bearing a proportionate share of total costin other words, the marginal cost of added shipments is in reality average cost of all shipments. Commonly, costs per unit of outputor variable costsare calculated by a formula of the type Y=a+bX, where Y is the ratio of total operating expenses to miles of track; a is "threshold" cost of operation, non-fixed investment which nevertheless must be made in some minimum quantity when any amount of activity is undertaken (such as the pay of one secretaryyou can't hire one-half a secretary); b is cost per unit of output; and X is the ratio of gross ton-miles of traffic to total miles of track. From the above formula, the I.C.C. derives what it calls a "percent variable." This is an expression of the proportion of variable costs to total costsbX. --- Y
Meyer et al., supra, at 274.
To estimate the costs of a particular service, the following procedure is used. Some costs, such as right-of-way maintenance, are developed for the system as a whole, on a gross ton-mile basis, and this average then applied to the movements in question. Other costs, capable of direct observatione. g., use of switch engines, special equipmentare then added, after application of the "percent variable" derived earlier. The result is the "out-of-pocket" cost of service. See, e. g., Rationale of Cost Finding Section, at R. 54 and following.
For our purposes, it is important to realize what kinds of railroad expenses are attributed to a service by the I.C.C. in calculating "out-of-pocket" costs. (These expenses are included by attribution whenever they are not susceptible of direct observation.) Total operating expense "includes all the labor, fuel, and miscellaneous variable costs associated with the operation of trains, yards, and stations; it also encompasses marketing, advertising, selling and promotion expenses under the catch-all heading of traffic expenses; even supervisory and legal expenses are included; furthermore, the major portion of capital consumption costs is found in the maintenance-of-way and structure and in maintenance-of-equipment accounts. In short, railroad operating expenses include virtually all important railroad costs except property taxes and interest payments." Meyer, et al., supra, at 275. Furthermore, shares of these tax and interest payments, as well as a return figure, are then assigned to the particular traffic under investigation in order to calculate "out-of-pocket" costs. See R. 55.
"Fully-distributed" costs include, in addition to "out-of-pocket" costs: the non-varying portion of operating expenses (b in the formula); the remainder of taxes and interest; the remainder of capital costs; and a 4% return on investment. "Fully-distributed" costs of a particular service are then obtained by adding an aliquot share of the system's fully-distributed cost components to the calculated "out-of-pocket" cost of the service. Cf. R. 55-56.NOTES
 References in this opinion to the report will employ the pagination of 313 I.C.C., thus: "R. 1 (2, 3, etc.)."
 This report embraced 43 docket proceedings.
 Its report was adhered to by five Commissioners. Commissioner Hutchinson concurred on the ground that the proposed schedule constituted a destructive competitive practice. Commissioner Freas, in a dissenting report in which Chairman Winchell and Commissioner Webb joined, dissented on the ground that the Interstate Commerce Act as amended neither required nor permitted "blanket protection for the water carriers." Commissioner McPherson, concurring in part, said: "I would approve all the rates which are compensatory but on this record I would not impose any differential." Only ten Commissioners were then in office.
 Flatcars on lease by the railroads capable of carrying two trailers.
 The conventional railroad-owned cars are generally capable of carrying only one trailer. As the Commission observed, the choice between 1- or 2-trailer cars "would rest entirely with the railroads."
 These six rates the railroads withdrew and they are not in issue in this litigation.
 The report makes it plain that the Commission considered that rates yielding in excess of out-of-pocket costs were "compensatory."
 The Commission pointed out that both the TOFC and the Sea-Land service were subject to certain variables not measurable in the record before it.
 The Commission, after describing the various competing services, concluded that "the preponderance of the testimony on these records is to the effect that most of the shippers prefer rail service to sealand except at lower rates for the latter." This, obviously, indicates recognition of some quality-of-service superiority in dependability and speed for overland as against water transport. R. 44. See also R. 38-40.
 The Congressional declaration of the National Transportation Policy was introduced into the Interstate Commerce Act by the Transportation Act of 1940, 54 Stat. 899, and now appears in the United States Code Annotated preceding §§ 1, 301, 901, and 1001 of Title 49:
"It is hereby declared to be the national transportation policy 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; to promote safe, adequate, economical, and efficient service and foster sound economic conditions in transportation and among the several carriers; to encourage the establishment and maintenance of reasonable charges for transportation services, without unjust discriminations, undue preferences or advantages, or unfair or destructive competitive practices; to cooperate with the several States and the duly authorized officials thereof; and to encourage fair wages and equitable working conditions;all to the end of developing, coordinating, and preserving a national transportation system by water, highway, and rail, as well as other means, adequate to meet the needs of the commerce of the United States, of the Postal Service, and of the national defense. All of the provisions of this Act shall be administered and enforced with a view to carrying out the above declaration of policy."
 See Dixie Carriers v. United States, 351 U.S. 56, 59, 76 S. Ct. 578, 580, 100 L. Ed. 934 (1956) ("`lower cost of equipment, operation, and therefore service'"; Congress' fear was lowering of rates by "strong" carriers, putting more efficient carriers out of business, citing 84 Cong. Rec. 5874); Statements of Chairman Freas before the Senate Committee on Interstate and Foreign Commerce, Hearings on S. 3778, 85th Cong., 2d Sess. (1958) ("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."); Colloquy between Senators Kefauver and Smathers, 104 Cong.Rec. 10859 (1958) ("Mr. Kefauver: * * * Some people have expressed the belief that under [15a(3)] it would be possible for one type of carrier to lower its rate to such an extent that another carrier would not be able to compete fairly on the basis of charging the overhead to that other carrier. There is nothing in [15a(3)], is there, which would enable one carrier to take undue advantage of another carrier * * *?" "Mr. Smathers: No. The answer is no.")
It will be observed that Congress had very different ideas as to what out-of-pocket and fully-distributed costs are than did the Commission. If Congress had realized that "railroad operating expenses include virtually all important railroad costs except property taxes and interest payments," see Appendix A, infra, it might have acted differently. But that, of course, is not a matter for our decision.
 That the Commission considered itself vested with a power of such sweeping breadth is further demonstrated by its stated conclusion herein (R. 50) that the water carriers must be favored by a differential under railroad boxcar rates, albeit a somewhat smaller differential than the 6% differential required of TOFC rates. Yet as to boxcar costs the Commission had found only that on some of this traffic Sea-Land "is shown as the low-cost agency; on the other traffic, the railroads' costs appear to be lower." R. 46. This surely falls far short of a finding that Sea-Land's is the low-cost mode.
 Locklin, Economics of Transportation 158 (1954).
 These figures are from ICC, Bureau of Accounts & Cost Findings, Distribution of Rail Revenue Contribution by Commodity Groups1952, Table 12 (1955).