Public Service Commission of Utah v. United States/Opinion of the Court

This appeal presents another clash between state and federal authority in the regulation of intrastate commerce. The Public Service Commission of Utah and the Utah Citizens Rate Association, appellants, seek to set aside an order of the Interstate Commerce Commission entered in a proceeding under § 13(3) and (4) of the Interstate Commerce Act in which an increase in intrastate freight rates to the general level of interstate rates was granted to railroads operating in Utah. 297 I.C.C. 87. The principal contention here is that the evidence before the Commission was insufficient to support its ultimate finding that existing intrastate rates caused 'undue, unreasonable, and unjust discrimination against interstate commerce.' 297 I.C.C., at 105. A three-judge District Court found against appellants on this and all subsidiary issues. 146 F.Supp. 803. Upon direct appeal, 28 U.S.C. § 1253, 28 U.S.C.A. § 1253, we noted probable jurisdiction. 1956, 352 U.S. 888, 77 S.Ct. 128, 1 L.Ed.2d 84. Having concluded that certain findings of the Commission lack sufficient support in the evidence, we reverse the judgment of the District Court.

The action of the Commission was limited to freight rates on intrastate traffic in Utah. In Ex Parte No. 175 the Commission had increased interstate freight rates on a national basis by an aggregate of 15%. The appellee railroads applied to the Public Service Commission of Utah for a like increase in intrastate rates. After a full hearing, the Utah Commission dismissed the application on the ground that the railroads had not produced evidence concerning their intrastate operations as required by Utah law. No appeal was taken. Instead, pursuant to 49 U.S.C. § 13(3) and (4), 49 U.S.C.A. § 13(3, 4), the railroads filed a petition with the Interstate Commerce Commission which led to the order under attack here. The Commission found the evidence insufficient to establish any undue or unreasonable advantage, preference, or prejudice as between persons or localities in intrastate commerce, on the one hand, and interstate commerce on the other. But in findings patterned after those approved in King v. United States, 1952, 344 U.S. 254, 73 S.Ct. 259, 267, 97 L.Ed. 301, it concluded that the intrastate rates caused 'undue, unreasonable, and unjust discrimination against interstate commerce.' 297 I.C.C., at 105. It sought to remove this burden by generally applying to intrastate traffic the 15% interstate increase previously granted in Ex Parte No. 175.

Appellants attack two findings of the Commission as not being supported by substantial evidence. The first is that existing intrastate rates were abnormally low and failed to contribute their fair share of the revenue needs of the railroads. Evidence was introduced to show that some of Utah's intrastate rates were lower than corresponding interstate rates for like distances. No showing was made, however, of the comparative costs of performing such services. The second finding under attack is that the conditions incident to intrastate transportation were not more favorable than those incident to interstate movements. The evidence underlying this finding indicated only that goods moving intrastate were handled precisely as were those in interstate transportation, being intermingled on the same trains.

Intrastate transportation is primarily the concern of the State. Federal power exists in this area only when intrastate tariffs are so low that an undue or unreasonable advantage, preference, or prejudice is created as between persons or localities in intrastate commerce on the one hand and interstate commerce on the other, or when those rates cast an undue burden on interstate commerce. Proof of such must meet 'a high standard of certainty,' Illinois Central R. Co. v. Public Utilities Comm'n, 1918, 245 U.S. 493, 510, 38 S.Ct. 170, 176, 62 L.Ed. 425; before a state rate can be nullified, the justification for the exercise of federal power must 'clearly appear.' State of Florida v. United States, 1931, 282 U.S. 194, 211-212, 51 S.Ct. 119, 123-124, 75 L.Ed. 291. The Court pointed out in State of North Carolina v. United States, 1945, 325 U.S. 507, 511, 65 S.Ct. 1260, 1263, 89 L.Ed. 1760, that the findings supporting such an order of the Interstate Commerce Commission must encompass each of the elements essential to federal power. Thereafter, in King v. United States, supra, we stressed the necessity of substantial evidence to support the findings, although we held it unnecessary 'to establish for each item in each freight rate a fully developed rate case.' 344 U.S. at page 275, 73 S.Ct. at page 270. In King, however, the insufficiency of the findings rather than of the evidence was urged upon the Court. Those findings, which we held adequate to support an order increasing intrastate rates, were, inter alia, (1) that existing intrastate rates were abnormally low and did not contribute a fair share of the railroads' revenue needs; (2) that conditions as to the movement of intrastate traffic were not more favorable than those existing in interstate commerce; (3) that the rates cast an undue burden on interstate commerce; (4) that the increase ordered by the Commission would yield substantial revenues; and (5) that such increase would not result in intrastate rates being unreasonable and would remove the existing discrimination against interstate commerce. 344 U.S. at pages 267 268, 73 S.Ct. at pages 266-267, footnote 13. We also held in King that the Commission might give weight to deficits in passenger revenue when prescribing intrastate freight rates so as to meet overall revenue needs. In our most recent review of federal power in this intrastate area, Chicago, M., St. P. & P.R. Co. v. State of Illinois, 1958, 355 U.S. 300, 78 S.Ct. 304, 2 L.Ed.2d 292, we relied on the principles of the above cases in striking down an increase in intrastate passenger fares for a suburban commuter service because the Commission had failed to take into account 'the carrier's other intrastate revenues from Illinois traffic, freight and passenger.' 355 U.S. at page 308, 78 S.Ct. at page 309.

We do not believe that the evidence here met the exacting standards required by our prior cases. As to the finding that prevailing intrastate rates were abnormally low and failed to contribute a fair share of overall revenue, we discover no positive evidence to indicate that the relative cost of intrastate traffic was as great as that of interstate shipments. The absence of such evidence is important, for it is not enough to say that interstate rates were higher on similar shipments for like distances, State of Florida v. United States, supra, 282 U.S. at page 212, 51 S.Ct. at page 124, especially where, as here, there was some indication that intrastate traffic moved at lower cost than interstate. The annual reports of the four interstate railroads operating in Utah showed that their Utah operating ratios (freight service cost divided by freight service revenue) and the Utah density statistics (ton mils of traffic per mile of main track) were more favorable than comparable system-wide figures. The Commission discredited the density statistics because of the absence of branch-line inclusion in the totals. This was true, however, in the case of both Utah and syste-wide computations, leaving no apparent foundation for the conclusion of unreliability.

Other evidence seemed to indicate that those railroads with the larger percentages of total operations within Utah enjoyed higher rates of overall return for 1953, the year just prior to the hearings in this case. The Denver & Rio Grande, with almost half of its entire operations within Utah, showed a rate of return of 6.06%. The Southern Pacific and Union Pacific, with substantially smaller proportions of Utah operations Showed returns of 3.48% and 3.56%, respectively.

Statistics introduced by the railroads as to comparative economic conditions showed recent economic improvement to be greater percentagewise in the West and particularly in Utah than in other sections. The emphasis recently has switched from agriculture to industrial and mining activity, with its resulting increase in traffic-a factor tending to suggest more favorable railroad operating conditions.

As to the finding that intrastate conditions were not more favorable than those incident to interstate transportation, the railroad evidence on this point was far from substantial. In essence, it merely showed that intrastate and interstate traffic was handled by the same crews and intermingled in the same movement. This evidence failed to establish that all material factors bearing on the reasonableness of rates were substantially the same. As we have previously noted, appellants offered convincing evidence not only of greater density on intrastate operations, permitting a wider spread of fixed costs, but also of lower operating ratios and higher returns as the percentage of intrastate traffic increased. In the face of this proof the evidence as to general similarity of conditions falls short of the 'high standard of certainty' required.

It is suggested that the Commission, in granting general interstate increases, frequently proceeds on the assumption that intrastate rates will be raised to the same level. But this assumption is no through ticket permitting it to approach the question of intrastate rates with partiality for a uniform increase. Rate uniformity is not necessarily the goal of federal regulation, nor can the Commission's wishful thinking be substituted for substantial evidence. Section 13 is not cast in terms of 'assumption' or 'partiality.' As applied to this case, it contemplates an inquiry into intrastate rates and conditions within Utah, and any conclusion that interstate operating conditions equally exist there must be ticketed on more than assumption.

Finally, we note an absence in the findings of any indication that the Commission concerned itself with the revenues derived from, or the conditions incident to, intrastate passenger operations. While a sweeping inquiry into those operations is not required, we believe that in light of our opinion in Chicago, M., St. P. & P.R. Co. v. State of Illinois, supra, the findings must reflect consideration of these factors in arriving at a general intrastate freight level. 'A fair picture of the intrastate operation, and whether the intrastate traffic unduly discriminates against interstate traffic, is not shown * *  * by limiting consideration to the particular *  *  * service in disregard of the revenue contributed by the other intrastate service.' 355 U.S. at page 308, 78 S.Ct. at page 309. This issue was not argued by the parties, our opinion in that case having been announced after submission of the instant case. We mention it at this point, however, since further proceedings before the Commission no doubt will ensue.

The judgment of the District Court is reversed and the cause is remanded to that court with instructions to set aside the order of the Commission and remand the cause to the Commission for further proceedings in conformity with this opinion.

It is so ordered.

Reversed and remanded with instructions.

Mr. Justice FRANKFURTER, whom Mr. Justice BURTON, Mr. Justice HARLAN, and Mr. Justice WHITTAKER join, dissenting.

This case involves an order of the Interstate Commerce Commission raising rates on Utah intrastate freight traffic on the ground that such rates unduly discriminate against interstate commerce. The Court has found the evidence insufficient to support the Commission's findings bearing on discrimination, and, although petitioners do not call them in question, has also concluded that the findings themselves are inadequate. Our holding in the recent case of Chicago, M., St. P. & P.R. Co. v. State of Illinois, 355 U.S. 300, 78 S.Ct. 304, 2 L.Ed.2d 292, has been extended to require, even in such a case as this, comprehensive findings concerning all intrastate operations.

There comes a time in the development of law, especially when it concerns as complex and important a subject as that in the present case, when a comprehensive survey must be made and the cumulative effect of episodic instances appraised to determine whether or not they reveal a harmonious whole. Case-by-case adjudication, without scrutiny of underlying generalizations or presuppositions, must culminate in an effort to determine whither we are going, and whether the direction cut by the specific instances should be further pursued or whether it represents a deviation from the path demanded by the purpose of the regulatory legislation. These considerations, and the fact that the sufficiency of the evidence can only be judged intelligently in the light of the findings, make it necessary to consider at some length the cases in which this Court has been concerned with intrastate rate discrimination against interstate commerce and the findings that have been required of the Commission to justify an order removing such discrimination.

Federal regulation of intrastate rates originated in cases involving discrimination as between particular persons or localities engaged in interstate commerce and particular persons or localities engaged in intrastate commerce. The discrimination in the Shreveport case, Houston, E. & W.T.R. Co. v. United States, 234 U.S. 342, 34 S.Ct. 833, 58 L.Ed. 1341, arose from the fact that interstate shippers were required to pay more than intrastate shippers although the rate disparity was not justified by any difference in costs, transportation conditions, or other factors usually considered in setting rates. This discrimination was removed in order to protect the interstate commerce of the particular shippers or localities shown to have been prejudiced, the same reason for prohibiting under § 3(1) of the Interstate Commerce Act, 24 Stat. 379, 380, as amended, 49 U.S.C. § 3(1), 49 U.S.C.A. § 3(1), any 'undue or unreasonable preference or advantage to any particular person * *  * locality *  *  * or any particular description of traffic *  *  * ' where two forms of interstate commerce are involved. It has often been said that this form of discrimination arises simply from the relation of rates to each other, but this is true only if it is understood that the circumstances that enter into the setting of rates and justify differences between them are also taken into consideration. Even in a persons-localities case, discrimination is not made out merely from a disparity in rates, but, as the Court made clear in the Shreveport case, the disparity must exist under "substantially similar conditions and circumstances * *  * ." 234 U.S. at page 347, 34 S.Ct. at page 834.

The power to remove discrimination thus announced in the Shreveport case, as between persons and localities in interstate and intrastate commerce, was given express statutory basis by § 416 of the Transportation Act of 1920, 41 Stat. 456, 484, 49 U.S.C. § 13(4), 49 U.S.C.A. § 13(4), amending § 13 of the Interstate Commerce Act. The amendment added, however, as part of a much more comprehensive regulation of the Nation's transportation system, a broad prohibition against 'any undue, unreasonable, or unjust discrimination against interstate or foreign commerce * *  * .' This provision, taken in conjunction with § 15a(2), 41 Stat. 456, 488, as amended, 49 U.S.C. § 15a(2), 49 U.S.C.A. § 15a(2), was construed in Railroad Commission of State of Wisconsin v. Chicago, B. & Q.R. Co., 257 U.S. 563, 42 S.Ct. 232, 66 L.Ed. 371, to authorize the Commission to remove 'revenue discrimination' resulting from unjustifiably low intrastate rates. The constitutionality of this power was upheld against claims of unwarranted intrusion into the area reserved to the States in their control over intrastate commerce. 'Revenue discrimination' consists, not in prejudice to particular shippers or localities, but in the burden cast on interstate commerce because of the failure of intrastate commerce to contribute its fair share to meet the revenue needs of the carrier. Of course interstate shippers and localities will ultimately be prejudiced by having to make up revenues properly due from intrastate commerce or by the collapse of an adequate transportation system, but the immediate purpose of the exercise of the Commission's power under this head is to assure to the carrier needed revenues. The test of what revenues are needed is found in § 15a(2), 41 Stat. 456, 488, as amended 49 U.S.C. § 15a(2), 49 U.S.C.A. § 15a(2), which instructs the Commission in prescribing rates to give due consideration to the need 'in the public interest, of adequate and efficient railway transportation service at the lowest cost consistent with the furnishing of such service; and to the need of revenues sufficient to enable the carriers, under honest, economical, and efficient management to provide such service.' The National Transportation Policy of 1940, 54 Stat. 899, 49 U.S.C. at pp. 7107-7108, 49 U.S.C.A. preceding section 1, in turn expanded and supplemented the test of § 15a(2). The 'dovetail relation,' as Mr. Chief Justice Taft termed it in the Wisconsin case, 257 U.S. at 586, 42 S.Ct. at page 236, between the second part of § 13(4) and § 15a(2), thus gave a much broadened purpose to federal regulation. 'Theretofore, the effort of Congress had been directed mainly to the prevention of abuses; particularly those arising from excessive or discriminatory rates. The 1920 Act sought to ensure, also, adequate transportation service.' Akron, C. & Y.R. Co. v. United States (New England Divisions Case), 261 U.S. 184, 189, 43 S.Ct. 270, 273, 67 L.Ed. 605.

The order sustained in the Wisconsin case affected intrastate fares on a state-wide basis. The Court expressly rejected discrimination against persons or localities as justification for the order, on the ground that there was insufficient evidence to support a finding of state-wide discrimination of this kind. This conclusion may have been justified in the Wisconsin case itself. However, the limitation imposed by that case and later cases upon the effective scope of the persons-localities basis of discrimination, because of stringent evidentiary requirements, seems to have derived more from the origin of the prohibition in the particular kind of situation presented by the Shreveport case than from restrictions inherent in the regulatory scheme contemplated by the statute. Whether discrimination is directly against shippers or against the revenues of the carriers, practical considerations require that evidence typical of discriminatory conditions be sufficient to justify an order that goes beyond the particular instances shown. In revenue cases evidence of some intrastate rates may be deemed typical and relied on to show that intrastate rates in general do not contribute their fair share to revenues. Likewise it might not have been unreasonable to rely on evidence of prejudice to certain interstate shippers as indicative of prejudice throughout the State to interstate shippers, and thus justify a state-wide order to remove discrimination against 'interstate commerce' without resort to the question of revenues.

For a time it appeared that § 13(4) would be construed to authorize the removal of discrimination against interstate commerce on such a basis. In Georgia Public Service Commission v. United States, 283 U.S. 765, 51 S.Ct. 619, 75 L.Ed. 1397, for example, the Court sustained a state-wide order not confined in its effect to traffic directly shown to have been prejudiced, yet attention does not seem to have been directed to the revenue needs of the carriers. In its opinion the Court spoke particularly of 'undue prejudice and discrimination to interstate shippers and localities * *  * ,' 283 U.S. at page 773, 51 S.Ct. at page 622 and the brief for the Interstate Commerce Commission indicates that the case was not primarily conceived of as a revenue case. Pp. 38 40. (See also the significance of the Court's reliance, 283 U.S. at page 774, 51 S.Ct. at page 622, on Nashville, C. & St. L.R. v. State of Tennessee, 262 U.S. 318, 43 S.Ct. 583, 67 L.Ed. 999.) In Louisiana Public Serv. Commission v. Texas & N.O.R. Co., 284 U.S. 125, 52 S.Ct. 74, 76 L.Ed. 201, an order affecting rates throughout most of the State was sustained in spite of the fact, as appellants pointed out, 284 U.S. at pages 126-128, 52 S.Ct. at page 75, that there were no findings in regard to revenue. Finally, in State of Ohio v. United States, 292 U.S. 498, 54 S.Ct. 792, 78 L.D. 1388, the Court sanctioned an unusually wide application of the persons-localities basis of discrimination in sustaining an order raising coal rates throughout the entire northeastern part of Ohio.

Nevertheless, the approach suggested by these cases was not in fact carried forward. The first Florida case, State of Florida v. United States, 282 U.S. 194, 208, 51 S.Ct. 119, 122, 75 L.Ed. 291, strongly reaffirmed the Wisconsin case in its rejection of the persons-localities basis for a state-wide order, and in the requirement that an order predicated on such a theory be restricted to 'competitive territory.' See American Express Co. v. State of South Dakota ex rel. Caldwell, 244 U.S. 617, 626, 37 S.Ct. 656, 661, 61 L.Ed. 1352. This was perhaps inevitable in view of the fact that the Shreveport doctrine had been evolved to remedy a specific situation-prejudice against particular shippers or localities shown be relatively direct evidence-so that the evil that gave rise to the legal theory in turn limited its growth. Furthermore, the broad power that the Court found in the relation between § 13(4) and § 15a(2), to protect the carriers' revenues, made it appear unnecessary to expand the persons-localities theory in order effectively to protect interstate commerce. The result has been that, over the years, the cases that have come before this Court that have resulted in significant developments under § 13(4) have been almost exclusively revenue cases.

Findings in revenue cases under section 13(4).-In these cases we have been concerned to set forth, with such precision as the subject matter permits, the findings required to justify an ultimate conclusion that there is a revenue discrimination against interstate commerce. Such findings need not be, in any partcular form of words, the automatic recitation of a talismanic formula, but must give ample assurance that the Commission has applied the standards and engaged in the process of judgment contemplated by the statute. Only through the findings can a court know what it is that the Commission has decided and is to be reviewed.

Since the purpose of the proceeding is to increase the contribution to revenues from intrastate traffic, there must be a finding that higher rates will in fact result in increased revenues. If business is unable to bear the higher rates, and either production will be curtailed or traffic diverted to cheaper modes of transport, increased rates may actually decrease revenues. The Commission's finding on this matter must express an informed, expert judgment about probabilities. It may not rest simply on the mechanical application of proposed rates to the volume of past traffic, but, on the other hand, the fact that there is uncertainty about the actual revenue outcome will not make the finding insufficient. United States v. State of Louisiana, 290 U.S. 70, 80, 54 S.Ct. 28, 33, 78 L.Ed. 181.

The finding principally required in a revenue case is that intrastate commerce is not contributing its 'fair share' to the revenues needed to achieve the ends set forth in § 15a(2) and the National Transportation Policy. What share is a fair share? Intrastate traffic is not contributing a fair share if it pays lower rates than interstate traffic when, on balance, the circumstances that usually go to the setting of rates are not substantially more favorable for intrastate traffic than for interstate traffic. As already indicated, this same test is applicable in cases involving discrimination against persons or localities. A disparity in rates does not alone establish a forbidden discrimination, as the Court has frequently said, nor, it is equally clear, since conditions surrounding interstate transportation may be more favorable, does identity of rates alone give assurance that there is not such discrimination.

The most obvious of the circumstances thus made relevant are the costs that enter into the rendering of the transportation service. The circumstances that go to the setting of a rate, however, are not necessarily confined to such costs, but may include a wide range of other considerations such as the ability of traffic to bear the increase and the economic condition of an industry. Costs, furthermore, may be considered indirectly through the factors that generate costs, such as the quality of the service rendered, the physical characteristics of the area through which the traffic moves, average loading, average length of haul, density of traffic, and so forth. In regard to the average length of haul, for example, if terminal costs are fixed, the longer the haul the more the carrier will realize per mile unless the rate is itself graduated to take this fact into account. Interstate hauls are probably on the whole longer than intrastate hauls. Where passenger service is concerned, wholly different factors may become important in deciding whether fare differentials are justified. For instance, coaches used in intrastate suburban or commuter service may require more cleaning and average fewer passenger miles per day than through coaches. See, e.g., In re Intrastate Rates Within Illinois, 102 I.C.C. 479, 483. It is apparent that the factors that may reasonably be found to enter into costs are inexhaustible in number and too changing in significance to permit a definitive catalogue to be drawn up. Least of all should such a task be undertaken by courts that cannot be assumed to have familiarity with, let alone specialized knowledge of, the practicalities of the transportation industry, and whose contact with these matters is necessarily episodic. Understanding in such a complex area as rate regulation, and the feel for the subject essential to successful administration, come only from saturation in the elements of the problem by those constantly concerned. Even to the Commission, what one year has appeared an inconsiderable factor in appraising relative transportation conditions affecting costs has the next, in the light of new experience, been deemed highly relevant.

It is essential to bear in mind precisely what it is that the Commission compares when it compares the circumstances surrounding the movement of interstate and intrastate traffic that go to the setting of rates. In the case of interstate traffic, one rate set for an entire interstate movement may take into account not only conditions within the State whose intrastate rates are in question, but also conditions in surrounding States. In comparing cost factors, therefore, it may be necessary to consider not only whether interstate and intrastate traffic as they move through the State move under the same conditions, but also the conditions under which interstate traffic moves beyond the borders of the State. The interstate area relevant to this inquiry may embrace several States, such as the Western District in the present case, or half the country; it is the area chosen in setting the interstate rates.

What findings the Commission is required to make concerning the circumstances surrounding interstate and intrastate transportation to justify an ultimate finding that intrastate traffic is not contributing its fair share to revenues have necessarily been dictated more by the practicalities of administration than the demands of abstract logic. The Commission has not been required to make findings as to revenues and costs attributable to intrastate traffic or even to make findings specifically negativing the existence of any factors that might affect costs. Although such findings would give increased assurance that intrastate commerce was not being made to bear more than its fair share, an unwillingness to render nugatory the provisions of § 13(4) by imposing upon the Commission obligations impossible of fulfillment has precluded the Court from such a requirement.

In Illinois Commerce Commission v. United States, 292 U.S. 474, 54 S.Ct. 783, 78 L.Ed. 1371, a case involving rates in the Chicago Switching District, the Court clearly laid down what had been implicit in earlier cases sustaining Commission orders. In reply to a contention that there had been no finding separating interstate and intrastate property, revenues, and expenses, the Court stated: 'Where the conditions under which interstate and intrastate traffic move are found to be substantially the same with respect to all factors bearing on the reasonableness of the rate, and the two classes are shown to be intimately bound together, there is no occasion to deal with the reasonableness of the intrastate rates more specifically, or to separate intrastate and interstate costs and revenues.' 292 U.S. at pages 483-484, 54 S.Ct. at page 787. This statement was quoted with approval and applied in King v. United States, 344 U.S. 254, 273, 73 S.Ct. 259, 269, 97 L.Ed. 301. The order in the King case was sustained although the only finding respecting the similarity of conditions surrounding interstate and intrastate transportation was the general finding that, 'the transportation conditions incident to the intrastate transportation of freight in Florida are not more favorable and such conditions in the Florida peninsula are somewhat less favorable than those (1) within southern territory and (2) between Florida and interstate points.' 278 I.C.C. 41, 72. See also the second Florida case, State of Florida v. United States, 292 U.S. 1, 11, 54 S.Ct. 603, 607, 78 L.Ed. 1077.

More elaborate findings have not been required because of the practical impossibility, at least where the traffic is mingled and carried on as one operation, of accurately segregating costs between interstate and intrastate commerce. Moreover, a general finding of similarity of conditions was held sufficient because it would have been the height of imprudence for this Court to require the Commission specifically to negative the existence of all factors that might touch on costs when, because of the changing nature of the transportation industry and the endless variety of situations giving rise to discrimination, such factors are not precise, fixed, or equal in importance. The holding in the King case was a practical solution hammered out to meet the almost intractable difficulty of regulating as two systems what is in fact one integrated transportation system of interstate and intrastate commerce, where regulation, if there is to be regulation at all, must be by approximation and compromise.

Evidence in section 13(4) cases.-In the present case the Commission found that, 'The conditions incident to the intrastate transportation of freight in Utah are not more favorable than those incident to the interstate transportation between Utah and adjoining States.' Petitioners attack this finding as not supported by substantial evidence. The findings necessary to support an order under § 13(4) have been considered at length because only in the light of them, and the reasons that led the Court to require these, and not more elaborate, findings, can the sufficiency of the evidence be appraised. Thus it would be paradoxical, after deciding that the Commission need not make findings segregating costs because of the practical impossibility of accurate allocation, to require that evidence to the same effect be in the record. Likewise, if individual findings on particular factors that may bear on costs need not be made because an authoritative catalogue of such factors cannot be compiled, it would be inconsistent, and would disregard the criteria governing the findings, to require that the record contain evidence as to all such factors or as to any particular factor.

Since, as cannot too often be pointed out, the Commission's administration of § 13(4) must from the nature of the case proceed by approximation and estimate, and for this reason the possible must have a large share in determining the permissible, it is highly relevant to examine at least some of the § 13(4) cases that have come before this Court to determine what evidence has actually been in the record in support of findings of similarity of conditions. In determining whether the record in the present case satisfied the requirement of substantial evidence, we do not deal with a new problem.

In the Wisconsin case itself, Railroad Commission of State of Wisconsin v. Chicago, B. & Q.R. Co., 257 U.S. 563, 42 S.Ct. 232, 66 L.Ed. 371, a stipulation that operating and transportation conditions were substantially the same made it unnecessary to examine the evidence underlying the Commission's finding on this point, and the record on which the Commission founded its order was not in fact before this Court. In State of New York v. United States, 257 U.S. 591, 42 S.Ct. 239, 66 L.Ed. 385, however, a case also involving intrastate passenger fares, this Court affirmed the District Court's dismissal of a bill brought to annul the Commission's order when the substantiality of the evidence was squarely in issue. The record contained testimony that physical characteristics of interstate and intrastate service, type of equipment, running time, and accommodations were similar, and in addition there was testimony of the most general sort that transportation conditions within the State did not justify lower rates. There was some indication that traffic density in New England was greater than in New York.

State of Alabama v. United States, 283 U.S. 776, 51 S.Ct. 623, 75 L.Ed. 1406, involved intrastate rates on fertilizer and fertilizer materials. The issue of the substantiality of the evidence to support the Commission's findings was squarely presented. The record contained no evidence of the cost of the intrastate service. Instead there was general testimony that the conditions surrounding intrastate and interstate transportation in Alabama were substantially similar, and that the witnesses knew of no conditions in Alabama that would justify different intrastate rates. There was some evidence that the tonnage of fertilizer moving intrastate was much greater than that moving interstate, and that ton-miles per loaded car of fertilizer were greater in Alabama than in the Southern District as a whole. Evidence was introduced of competition from wagons on short, intrastate hauls. Yet in spite of this specific evidence tending to show, as against the more general testimony, that conditions affecting intrastate traffic were not the same as those affecting interstate traffic, this Court sustained the Commission's order, finding that the question of similarity of transportation conditions had been 'thoroughly canvassed,' and that the findings were supported by evidence. 283 U.S. at page 779 and note 4, 51 S.Ct. at page 624.

In Louisiana Public Service Commission v. Texas & N.O.R. Co., 284 U.S. 125, 52 S.Ct. 74, 76 L.Ed. 201, involving both interstate and intrastate rates on sand, gravel, and similar materials in western Louisiana, this Court found that the facts stated in the opinion were adequately supported by the evidence and were sufficient to warrant the order prescribing higher rates. The record contained evidence as to general freight costs per mile in different States in the territory embracing Louisiana, but no comparison was made of the costs of interstate and intrastate transportation within the State, and there was no evidence of the cost of transportation in the particular part of Louisiana concerned or of the particular commodities on which higher rates were sought. Evidence was introduced to show general freight density in the States in the territory considered, interstate and intrastate combined, and, according to the Commission, 155 I.C.C. 247, 253, about ninety per cent of the movements of the particular commodities involved was intrastate. There was also considerable evidence that bore indirectly on traffic density in these commodities: tonnage moved in each of the States; producers throughout the territory, in Louisiana, and in western Louisiana; total production in western Louisiana; shipments to particular consumers; estimates, based on the Louisiana highway program and population statistics, of probable future consumption. In addition there was the usual general testimony on similarity of conditions, and references to Commission findings in earlier proceedings on the same question. Although the record is, as to the issue we are here concerned with, one of the most extensive in any § 13(4) case that has been before the Court, there is little or no evidence specifically directed to transportation conditions in the southern part of western Louisiana, the area to which the Louisiana Commission had specifically refused to apply the interstate level of rates.

Another case in which this Court found that there was substantial evidence was State of Florida v. United States, 292 U.S. 1, 54 S.Ct. 603, 78 L.Ed. 1077, involving intrastate rates on logs. The record contained evidence bearing on the cost of transporting logs in Florida, and indicating that existing intrastate rates did not cover the cost of the service. As to freight in general, there was evidence that costs per gross ton-mile were higher on the carrier's lines in Florida than for the rest of its system, and that density on its lines in Florida, intrastate and interstate combined, was less than for its system as a whole or its system excluding Florida. These comparisons involved the question whether evidence of general freight conditions could be used to justify raising rates on a particular commodity. The case furnished the best evidence on the probability that increased rates would increase revenues because the proposed intrastate rates had actually been in effect for a period before the Court set aside an earlier order of the Commission.

In Illinois Commerce Commission v. United States, 292 U.S. 474, 54 S.Ct. 783, 78 L.Ed. 1371, already referred to in connection with the rule that there need not be findings segregating interstate and intrastate expenses, the Court also found that there was substantial evidence. The Commission relied primarily on an extensive cost study of the movement of all commodities in the Chicago Switching District, a study that did not, however, make any separation between interstate and intrastate costs. There was evidence of the physical characteristics of the service: that interstate and intrastate traffic were handled in the same manner and carried on the same trains. On the other hand, it was shown that average intrastate hauls were shorter than average interstate hauls, yet the Commission had set a single rate no matter what the length of haul. In spite of this evidence indicating that at least one factor underlying costs favored intrastate traffic, the Court upheld the Commission's order.

State of Mississippi ex rel. Rice v. United States, 307 U.S. 610, 59 S.Ct. 822, 83 L.Ed. 1493, concerned intrastate rates on sand, gravel, fertilizer, and fertilizer materials. This Court affirmed the District Court per curiam, citing cases to the effect that the Commission's orders would be sustained when supported by substantial evidence. The record contained no evidence segregating costs between intrastate and interstate traffic, although one witness testified generally that costs were less in Mississippi than in neighboring States. There was no evidence on traffic density as such, but considerable from which conclusions as to density might be inferred. Thus there was evidence of the tonnage of the commodities involved that had been moved by the carriers, intrastate, interstate, and total, during selected annual periods, and evidence of points of production in Mississippi. The carriers relied on testimony that the same service was rendered interstate and intrastate traffic and the usual general evidence of similarity of conditions. Against this was testimony that the bulk of intrastate movements were single-line, and that, because of extra switching and inspection of cars at switching points, and because of the necessity of keeping interchange records, joint-line movements were more expensive than single-line. Dispute centered principally on whether increased rates would increase revenues. Evidence was introduced to show intensive truck competition, facilitated by the fact that short hauls from wayside gravel pits to all points in the State were possible. Other evidence minimized the seriousness of such competition. The record contained little evidence of conditions surrounding the production and transportation of the commodities in other parts of the South, but the Commission relied to some extent on its investigation into general conditions in earlier proceedings concerned with interstate rates.

State of New York v. United States, 342 U.S. 882, 72 S.Ct. 152, 96 L.Ed. 662, affirming, D.C., 98 F.Supp. 855, another per curiam affirmance, raised intrastate commuter fares to the level of interstate commuter fares. A study had been made of the cost of the intrastate commuter service, costs being apportioned between that service and other service on the trains studied principally on the basis of passenger-mile ratios. The results of the study were used by the Commission, however, to show that intrastate fares did not cover the cost of service, rather than to contrast intrastate with comparable interstate costs. Intrastate commuter service was of much heavier volume than interstate, both in number of passengers and passenger miles, but the Commission pointed out that increased costs from such density went far to offset its advantages. Wear and tear on equipment was increased, and since traffic was concentrated at rush hours, additional crews and equipment needed to handle it were idle during the rest of the day. There was evidence that intrastate trains made more stops than interstate trains and used coaches that were older, more crowded, and not air-conditioned. Yet these differences were found not to justify a lower fare.

In King v. United States, 344 U.S. 254, 73 S.Ct. 259, 97 L.Ed. 301, the parties raised no question as to the sufficiency of the evidence. The Court observed, however, that evidence supporting the findings appeared in the record and that much of the material that had been before the Commission in its investigations into the interstate rates had also been before it in the § 13 proceedings. 344 U.S. at page 272, 73 S.Ct. at page 269. The record did contain evidence of operating expenses allocated to Florida traffic, interstate and intrastate, but it was admitted that the allocation was simply on the basis of the percentage of revenue derived from such traffic, and that it was impossible accurately to ascertain actual expenses. However, for one carrier, because of special bookkeeping, it was possible to show that actual freight expenses were higher in Florida than for the rest of its system. Statistics were introduced to show that the percentage increase in net railway operating income for Florida and the carriers operating in Florida had been greater than for the Southern District as a whole.

As to density, there was evidence that the density of freight traffic in Florida, at least for some carriers, was lower than that for their whole systems or their systems excluding Florida. Against this the State Commission introduced evidence showing that there had been a greater relative increase in tonnage originated in Florida than on the entire inated in Florida than on the entire systems of the three principal carriers serving Florida. There was evidence that intrastate movements were mostly by local trains on branch lines, and the percentage of branch lines in Florida was greater than for the rest of the carriers' systems. Because of overtime wages, shorter hauls, more switching, the necessity of frequently making and breaking up trains, such local trains were shown to be more expensive to operate than through trains. The Court left it to the Commission to weigh the competing claims of all this evidence on the question of similarity of conditions.

In recent years the Court has, by per curiam affirmance, disposed of a number of cases involving challenges to § 13(4) orders. In State of Tennessee v. United States, 346 U.S. 891, 74 S.Ct. 222, 98 L.Ed. 394, affirming, D.C., 113 F.Supp. 634, a case involving intrastate rates on coal and wood, the District Court had found that there was substantial evidence. The Commission's report, Tennessee Intrastate Rates and Charges, 286 I.C.C. 41, stated the principal evidence relevant to the finding that intrastate conditions were not more favorable than interstate conditions. The results of a traffic study, separately listing interstate and intrastate terminations of the commodities involved, bore indirectly on density. A greater proportion of intrastate than interstate traffic was handled in costly local trains. A cost study based on waybills showing average load, haul, and rates charged was introduced to show that intrastate rates provided a greater return above cost than rates on interstate movements; the Commission rejected this evidence on the ground that the information in the waybills was unreliable.

In Louisiana Public Service Commission v. United States, 348 U.S. 885, 75 S.Ct. 206, 99 L.Ed. 695, affirming, D.C., 125 F.Supp. 180, the District Court had also found that there was substantial evidence. The Commission in its report, Louisiana Intrastate Freight Rates and Charges, 291 I.C.C. 279, relied on evidence that generally intrastate traffic was carried in the same trains with the same crews as interstate traffic and under no more favorable operating conditions, and that in fact considerable intrastate traffic moved in expensive local trains. There was also evidence separately stating, as between intrastate and interstate commerce, the tonnage of the various commodities terminated during a test period.

In Illinois Central R. Co. v. Mississippi Public Service Commission, 349 U.S. 908, 75 S.Ct. 599, 99 L.Ed. 1244, affirming Mississippi Public Service Commission v. United States, D.C., 124 F.Supp. 809, the principal question before the District Court had been the sufficiency of the evidence. Among the considerations that that court relied on in setting aside the Commission's order was the fact that the passenger deficit in Mississippi was lower than for the entire systems of the Mississippi carriers, for the rest of the South, or for the country as a whole. It found the evidence too unsubstantial, furthermore, to support the Commission's judgment that, in spite of competition from other forms of transportation, increased rates would increase revenue.

From this review of the cases in this Court, and from a consideration of others in the District Courts and before the Commission, certain conclusions emerge that should be decisive in disposing of the case now before us. In the first place, there has been in many cases only the slightest direct evidence of the cost of moving intrastate traffic, and in other cases the record is wholly devoid of such evidence. As is clear from Illinois Commerce Commission v. United States, 292 U.S. 474, 54 S.Ct. 783, 78 L.Ed. 1371, and King v. United States, 344 U.S. 254, 73 S.Ct. 259, 97 L.Ed. 301, such evidence is not required when intrastate and interstate traffic are intimately bound together. The records in many cases contained evidence of factors affecting costs, but, according to the traffic involved and the character of the investigation, what has been deemed an important factor in one case has been passed over in silence in another. Reliance on general testimony concerning similarity of conditions has been almost universal and, in some of the cases considered, such evidence appears to have provided the principal, if not exclusive, support for the Commission's finding.

The evidence introduced before the Commission to support a finding of similarity of conditions varies with the purpose and scope of the proceeding. Thus, an investigation under § 13(4) may involve intrastate rates on a single commodity or on all freight traffic; it may involve the rate of a single carrier or all the carriers in the State; it may be confined to rates in a certain part of the State or extend to rates and fares throughout the State. Obviously a single, comprehensive, easily applied formula for testing the evidence required in all these cases cannot be devised without closing one's eyes to the rich and shifting variety of situations presented by the Nation's transportation system, and without unduly confining the Commission in its responsibility to deal with the intermingled interstate and intrastate transportation of a single system.

Finally, it is of importance to distinguish § 13 proceedings, such as those in the present case, that accompany or follow regulation of interstate rates on comparable traffic. In these cases, in determining the level at which interstate rates should be set in order to assure a given revenue, the Commission often proceeds on the assumption that intrastate rates will be raised to the same level or proportionately increased. Of course uniformity of rates is not the goal of federal regulation, and the Commission's conclusions in the interstate proceedings do not justify foregoing the inquiry into intrastate rates and conditions required by § 13. But the manner in which the Commission customarily proceeds is strong evidence of what is practically possible in performing the difficult regulatory task imposed by the statute. It should be respected by this Court in prescribing standards for the Commission's guidance. When a § 13 proceeding follows regulation of rates on comparable interstate traffic, furthermore, evidence introduced in the interstate proceedings, whether they are general revenue proceedings or directed to specific rates, will, to a greater or less extent, also bear on conditions surrounding the movement of intrastate traffic, and it may not be unreasonable for the Commission to assume, in the absence of persuasive evidence to the contrary, that the conclusions it has drawn from such evidence about general conditions are equally applicable to a particular State. As was stated in King v. United States, 344 U.S. 254, 272, 73 S.Ct. 259, 269, 97 L.Ed. 301, 'In the absence of any showing that it is not applicable to Florida, the evidence which forms the basis of the Commission's nationwide order becomes the natural basis for its Florida order.'

The evidence in the present case.-The present case concerns, with exceptions not now relevant, all Utah intrastate freight rates. It follows a general revenue proceeding raising interstate rates, and seeks to apply the same increase there granted to intrastate traffic.

The validity of the Commission's state-wide order was not impaired because as to particular traffic or carriers in the State a rate increase might not be justified. It may be that an increase in rates on certain traffic will decrease revenues, or the operations of one carrier may already be exceptionally profitable. Factors affecting costs may vary throughout the State. In State of New York v. United States, 257 U.S. 591, 42 S.Ct. 239, 66 L.Ed. 385, the fact that evidence indicated that one carrier had a more favorable route with greater density than others did not preclude an order raising passenger fares on a state-wide basis. If general orders are to be possible at all, and § 13(4) necessarily implies them, the Commission must be able to proceed on evidence typical of general conditions, see Georgia Public Service Comm'n v. United States, 283 U.S. 765, 774, 51 S.Ct. 619, 622, 75 L.Ed. 1397, making provision as it did in the present case for the re-examination of specific rates claimed not to fall within the findings. See Railroad Commission of State of Wisconsin v. Chicago, B. & Q.R. Co., 257 U.S. 563, 591, 42 S.Ct. 232, 238, 66 L.Ed. 371; Akron C. & Y.R. Co. v. United States (New England Divisions Case), 261 U.S. 184, 196-199, 43 S.Ct. 27, 275-276, 67 L.Ed. 605.

The evidence in the present case said to bear on the conditions surrounding the movement of interstate and intrastate traffic included evidence of the following: economic conditions in Utah, the rate of return on the carriers' net property investment, net railway operating income, revenue per ton-mile, freight service ratio, density, operating efficiency, the character of the service rendered interstate and intrastate traffic, and general similarity of operating conditions.

There was general testimony from a number of qualified witnesses that interstate and intrastate traffic in Utah moved under substantially similar conditions, were carried on the same trains, handled by the same crews, and accounted for in the same manner, and that generally no better service was given interstate than intrastate traffic. There was evidence that so-called 'piggyback' service was used exclusively for interstate traffic, and that traffic in Utah was more heavily weighted in favor of mine products moving in cheap gondola cars than in neighboring States, but there was no indication that these considerations, insofar as they affected costs, were not in fact taken into account in setting rates on the particular commodities involved.

The average rate of return on net property investment for the five Class I carriers serving Utah, and their net railway operating income, were shown to have declined steadily since 1950. Since these figures were not broken down between interstate and intrastate operations within Utah, however, nor between operations in Utah and other States, but instead embraced the carriers' entire systems, the showing of a decline is of no probative value on similarity of conditions surrounding the movement of interstate and intrastate traffic. The decline may be attributable as much to one traffic as to the other. It goes instead to establish the carriers' need for additional revenue from one source or another. The same may be said of the evidence of particular costs entering into the rendering of freight service. There was evidence that wages on the entire systems of the five Utah Class I carriers had risen 45.89% from 1948 to 1953, that they were generally higher in the Western District than in the States further east, and that average unit prices for materials and supplies in that District had risen 24.7% from 1949 to 1954. In none of these statistics were costs attributable to interstate and intrastate traffic segregated and compared.

As in most § 13(4) cases, the record here contained numerous statistics showing revenue per ton-mile. Revenue per ton-mile on all freight traffic in Utah, interstate and intrastate, had increased less between 1939 and 1952 than revenue per ton-mile for the Western District and for the country as a whole. On the other hand, the results of a waybill study introduced by the Utah Commission showed that in 1952 the revenue per ton-mile from Utah intrastate shipments was greater than that from interstate shipments that had originated in Utah. But, as the Interstate Commerce Commission pointed out in its report revenue per ton-mile tends to be less on interstate traffic because average hauls are longer. In any event, since differences in revenue per ton-mile are often simply a reflection of the fact that there are differences between interstate and intrastate rates, they are not probative of the relative cost factors affecting interstate and intrastate traffic.

I turn now to the evidence that the Court finds, contrary to the Commission, establishes that on the whole conditions surrounding intrastate traffic are substantially more favorable than those surrounding interstate traffic, and require overruling the District Court and setting aside the Commissioner's order. This evidence, according to the Court, so preponderates over the general testimony on similarity of conditions, referred to above, that the Commission was bound to accept it as decisive.

There was evidence such as increase in population, automobile registrations, and income payments to individuals that justified a conclusion that economic conditions in Utah had improved relatively more than in the Western District and the United States as a whole. Although this Evidence may be relevant to the density of traffic in Utah as compared with other States, or the density of intrastate and interstate traffic within Utah, it is so remote that it cannot reasonably be argued that it compels a conclusion one way or another.

The Court suggests that Utah intrastate transportation conditions must be more favorable because the rate of return on the Denver and Rio Grande Western, with almost half of its operations in Utah, is greater than the rate of return for the Southern Pacific and Union Pacific. For this Court to draw such an inference presupposes a natural law of railroading within the Court's knowledge or to which it can gain ready access. There is nothing in the record to indicate that the higher rate of return for the D&RGW is due to the fact that a substantial part of its traffic is Utah intrastate traffic. It may be due to any one of a number of factors: the railroad as a whole may be more efficiently operated than the UP or the SP; its routes may be more favorably situated or the operations on other parts of its system particularly profitable; the lower rate for the UP and the SP may be due to circumstances present anywhere in their farflung empires.

The Court finds that there was no justification for the Commission's conclusion that density figures introduced by petitioners were unreliable. Such a judgment by the Court on the basis of these statistics is indeed puzzling.

Exhibit 64 compares, for four Class I carriers in Utah, freight density (ton-miles per mile of main-line track) for 'intrastate' operations with this density for each carrier's system as a whole. In each case the 'intrastate' density is greater than the system density. It is important to understand just what is being compared in this exhibit. As was admitted by petitioners, the 'intrastate' density in the exhibit is not limited to traffic that, under § 1(2)(a) of the Act, 24 Stat. 379, as amended 49 U.S.C. § 1(2)(a), 49 U.S.C.A. § 1(2)(a), moves 'wholly within one State.' Yet it is the cost of moving only such intrastate traffic, traffic wholly within the State, that must be ascertained and compared with the cost of moving interstate traffic, because only this kind of intrastate traffic is subject in the first instance to regulation by the State. Instead, 'intrastate' density in the exhibit includes also that portion of interstate movements that lies within Utah, whether such movements originate or terminate there or bridge the State.

The comparison in this exhibit, therefore, is by no means probatively the best when the purpose is to show that the cost of moving one ton one mile on an intrastate journey is less than it would be on an interstate journey. The most persuasive comparison would be between the average density of interstate traffic in the region taken for the purposes of setting the interstate rates and the average density of intrastate traffic that moves wholly within the State. The interstate density properly taken into consideration, because it is the density that affects the interstate rates, is the average density for the entire interstate journey including that part that lies in other States, but not including the intrastate density in those States. There is some, although less, probative value in a comparison between the density of interstate traffic confined to interstate traffic within the State and the density of intrastate traffic. For this comparison to be reliable, density of interstate traffic for the entire region for which the interstate rates have been set must be assumed to be substantially the same as the density of interstate traffic within the State.

The comparison in Exhibit 64, between interstate and intrastate density on the carriers' entire systems and interstate and intrastate density in Utah, is of considerably less probative value than these two comparisons. It does not show whether Utah intrastate density is in fact greater or less than interstate density either in Utah or on the carriers' entire systems. It has some evidentiary value, of course, since, if over-all system density is less than over-all density within the State, there are probably factors in the State generating traffic that do not exist in neighboring States. That these factors generate more intrastate than interstate traffic, however, is a matter of speculation, and average regional interstate density may still be higher than intrastate density within Utah. A comparison somewhat similar to that in Exhibit 64, and in fact relied on in the second Florida case, Georgia Pub. Serv. Comm'n v. Atlantic C.L.R. Co., 186 I.C.C. 157, 164, is a comparison between combined interstate and intrastate density within one State and combined interstate and intrastate density on the balance of the carrier's system in surrounding States. It is impossible to tell from such a comparison whether or not region-wide interstate density is less than intrastate density in the particular State involved, although that it is may be a permissible judgment for the Commission to make but hardly for this Court to impose upon it.

These distinctions are necessary because the bearing of density on the cost of interstate and intrastate traffic is unlike that of many other factors. Thus road conditions, wages, and cost of fuel in a State are likely to fall with equal effect on intrastate traffic and interstate traffic as it passes through the State. As to these conditions, a comparison between those prevailing in one State and surrounding States is highly relevant. Since interstate rates take into account average conditions throughout the region, if conditions are more favorable in one State, intrastate traffic moving only under those more favorable conditions should not have to pay the same rate as interstate traffic.

The Commission itself discounts the importance of the density figures because comparison is made of density on main-line track only and that on branch lines is excluded. However, the Court finds that, since branch-line density is excluded both from Utah and system figures, the omission does not render the evidence unreliable. This is right if, and only if, the proportion of branch line to main line in Utah is the same as for the carriers' systems as a whole. If there is proportionately more low-density branch-line track in Utah, Utah density will be relatively lower. The possibility that such is in fact the case was mentioned at the hearings (R. 294), and petitioners introduced no evidence to supply the deficiency in their exhibit.

Thus it is clear that there were good reasons for rejecting the evidence in Exhibit 64, but even if the comparison offered were the best, the Commission surely would not be compelled to accept the density evidence as conclusive. Density is only one of a multitude of factors affecting costs, and what is gained by high density may be lost because of some other factor. High density itself, as was shown in State of New York v. United States, 342 U.S. 882, 72 S.Ct. 152, 96 L.Ed. 662 (see 279 I.C.C., at 161), is not always an unmitigated blessing, and low interstate density may be the result of rate discrimination rather than a justification for it.

The Court also relies on evidence that the freight service operating ratio of four Class I carriers operating in Utah for 1950-1953 was more favorable for operations described in the exhibit as 'intrastate' than for the carriers' entire systems. The 'intrastate' traffic included in this comparison was the same as that used in the density statistics. Moreover, the Commission itself rejected the evidence because of an even more fundamental criticism of the method used to segregate expenses attributable to intrastate traffic. Since it was impossible in many instances to ascertain actual expenses, allocation had been to a large extent simply on the basis of the number of train miles in the State. The carriers had long attacked this method of allocation as artificial and not reflecting the actual cost of service, and there is no justification for overriding the Commission's judgment on the matter.

The upshot is that petitioners produced no evidence of dissimilar conditions that was not open to serious criticism and that the Commission was not justified in rejecting or severely restricting in effect. There remained to support the Commission's finding the general testimony that interstate and intrastate traffic in Utah did in fact move under substantially similar conditions, and also the evidence in the earlier general revenue proceeding to adjust interstate rates that, as we recognized in King v. United States, 344 U.S. 254, 272, 73 S.Ct. 259, 269, 97 L.Ed. 301, becomes the natural basis for the intrastate order in the absence of any showing that it is not applicable. This Court has never before overturned an order of the Commission under § 13(4) on the ground that the evidence was insufficient to sustain a finding of revenue discrimination, although that question has been squarely presented on a number of occasions. The evidence in the present case was as substantial as that in many of these other cases, and, under an interpretation of the Act that respects the practical difficulties facing the Commission, should be enough.

The Court objects that there was 'no positive evidence' to support the Commission's finding, but it does not say what evidence on what factors affecting costs, beyond that in the record, was required. Should there have been evidence on length of haul, average loading, cars per train, grade conditions, volume of intrastate traffic handled by local trains, size of crews? This Court cannot say in each case what factors would be significant enough to necessitate evidence, and to require that the record contain evidence negativing all possible factors would be as paralyzing as to require findings to that effect. Once there is evidence of the general nature here introduced, it is for those who contend that intrastate conditions are dissimilar to come forward with convincing evidence showing what specific factors affecting costs are more favorable. Statements in earlier opinions of the Court indicate that such is in fact present law. It is the only workable solution. As Mr. Justice Brandeis said in the Akron, C. & Y.R. Co. v. United States (New England Divisions Case), 261 U.S. 184, 197, 43 S.Ct. 270, 275, 67 L.Ed. 605, 'Obviously, Congress intended that a method should be pursued by which the task, which it imposed upon the Commission, could be performed.'

Intrastate passenger operations.-This consideration of the findings the Court has required in § 13(4) cases, and the evidence necessary to support them, bears on the far-reaching consequences, not adverted to in the Court's opinion, of holding that our decision in Chicago, Milwaukee, St. P. & P.R. Co. v. State of Illinois, 355 U.S. 300, 78 S.Ct. 304, 2 L.Ed.2d 292, in some manner applies to the present case. It is not clear from the Court's opinion whether on the remand the Commission will be required to make findings on the profitableness of intrastate passenger operations, such as were required in the Milwaukee case for all intrastate operations, or only on the similarity of conditions surrounding intrastate and interstate passenger traffic.

It was settled in King v. United States, 344 U.S. 254, 73 S.Ct. 259, 97 L.Ed. 301, that interstate freight traffic could be made to support interstate passenger traffic, and intrastate freight traffic to support intrastate passenger traffic. Left open, because there was no indication in the record that the deficits arising from the two kinds of passenger traffic significantly differed, was the question whether intrastate freight traffic could be made to support an interstate passenger deficit. Such support would be the practical effect if interstate freight rates are set to compensate for an interstate passenger deficit, and intrastate freight rates are raised to the same level although intrastate passenger operations are profitable or result in a smaller loss than interstate passenger operations. The question was presented in Mississippi Public Service Commission v. United States, D.C., 124 F.Supp. 809, where the failure of the Commission to take into account a lower passenger deficit in Mississippi was one ground for the District Court's setting aside the order. Since there were other issues in the case, however, our per curiam affirmance did not necessarily indicate a view on the question. Illinois Central R. Co. v. Mississippi Public Service Commission, 349 U.S. 908, 75 S.Ct. 599, 99 L.Ed. 1244. The precise problem, furthermore, was not clearly settled by the general holding in the Milwaukee case that there should be findings 'which reflect the commuter service deficit in the totality of intrastate revenues * *  * ,' 355 U.S. 300, 308, 78 S.Ct. 304, 309, 2 L.Ed.2d 292, nor, strictly speaking, must the question be answered at this point in the present case though doubtless it will arise under the Court's disposition on the remand.

There is no apparent reason why a lower passenger deficit, like any other favorable circumstance surrounding intrastate transportation, would not justify a lower rate on intrastate freight traffic. If intrastate traffic taken as a whole contributes its fair share to needed revenues and does not, from a revenue standpoint, discriminate against interstate commerce, what justification can there be for a finding of discrimination that is possible only because a segment of intrastate traffic is considered in isolation? Raising rates in this situation may have the effect of compelling intrastate commerce to contribute more than its fair share.

It does not necessarily follow that the Commission should be required to make findings, supported by evidence in the record, that the intrastate passenger deficit is not lower than the interstate, or about the profitableness of, or circumstances surrounding, any segment of intrastate operations with which it is not immediately concerned. Indeed, the consequences that follow in the train of such a requirement demonstrate that it exalts formal consistency over sound policy. In the first place, if the Commission must determine the profitableness of intrastate passenger operations, it will of course be compelled to segregate revenues and costs attributable to intrastate and interstate traffic. Yet it is precisely this that in Illinois Commerce Commission v. United States, 292 U.S. 474, 54 S.Ct. 783, 78 L.Ed. 1371, and the King case we said, urged on by the difficulty of accurate allocation, was not required when interstate and intrastate traffic were mingled together.

Another consequence of the Court's decision appears to be that every case involving an intrastate rate claimed to result in a revenue discrimination must be broadened into a general inquiry into all intrastate rates and the profitableness of, or circumstances surrounding, all intrastate traffic. The result would be a radical, and in all likelihood unworkable, change in the way the Commission has administered the provisions of § 13(4) for over 35 years. The Commission's Reports are full of cases in which intrastate rates on one commodity or group of commodities, or on traffic in only one part of a State, have been tested for discrimination without reference to the entire intrastate picture. The Wisconsin case itself raised passenger fares without consideration of other intastate operations, and the Chicago Switching case, Illinois Commerce Commission v. United States, 292 U.S. 474, 54 S.Ct. 783, 78 L.Ed. 1371, concerned only segments of Illinois and Indiana intrastate traffic. The possible disruptive effect of requiring the Commission to proceed by giant, state-wide strides, rather than by steps designed to relieve discrimination from a particular segment of intrastate commerce, is alone sufficient to cast doubt on the wisdom of the Court's decision and to require a close scrutiny of the Milwaukee case to determine if the rule there set forth is not in fact confined to the special situation that gave rise to it.

The Milwaukee case involved intrastate fares on commuter service that was for all practical purposes totally separate from the interstate operations of the carrier. There had been no previous proceedings to set the fares on comparable interstate traffic because in fact there was no significant comparable interstate traffic with which the intrastate traffic was mingled. Intrastate fares were not raised to a level of interstate fares that had previously been determined to be reasonable; instead, they were raised to a level that the Commission decided, after considering the intrastate traffic alone, would contribute a fair share to revenues. In such a context a determination of discrimination too easily becomes simply an inquiry into reasonableness, an inquiry the Commission is not empowered to make in respect to intrastate fares. By being compelled to make findings reflecting a broader view of the profitableness of intrastate operations, the Commission is made to give assurance that intrastate commerce has in fact been compared with interstate commerce and discrimination found. A broad comparison between interstate and intrastate operations is necessary because, in this special situation, a narrow comparison is not possible.

It is a very different matter when there is comparable interstate traffic and intrastate rates are raised to the level of rates on that traffic already determined to be reasonable. This was the case in State of New York v. United States, 342 U.S. 882, 72 S.Ct. 152, 96 L.Ed. 662, affirming D.C., 98 F.Supp. 855, also involving intrastate suburban fares, and there was no suggestion that the Commission was bound to look to the totality of intrastate operations. The fact that a comparison of rates on similar traffic is possible gives a greater degree of assurance than was possible in the Milwaukee case that intrastate traffic is not being compelled to contribute more than its fair share to needed revenues, and that it is discrimination rather than simply unreasonableness that the Commission seeks to remedy. In this situation the Commission is justified in considering whether a particular segment of intrastate commerce is contributing its fair share to revenues.

Of course, those who contend that intrastate traffic as a whole is not discriminating against interstate traffic may come forward and show, as they may in respect to any claimed dissimilarity of conditions surrounding interstate and intrastate traffic, some favorable aspect of intrastate operations that the Commission should take into account. In the absence of such a showing, however, the Commission should be able to assume that discrimination shown to exist as to the particular segments of intrastate and interstate traffic with which the § 13(4) proceeding is concerned is not offset by other conditions that this Court speculates may affect wholly different segments of intrastate commerce. The record in the present case is devoid of the remotest suggestion that the Utah intrastate passenger deficit is any less than the interstate passenger deficit, and the Commission should not be required to seek out such evidence itself and make findings beyond those it has already made.

This is a solution that accommodates imponderables and does not demand precision where the nature of the subject can yield only approximations. It is a solution responsive to the difficult regulatory problems posed by § 13(4). Embedded in it are some of the advantages in simplicity of administration that would follow if this Court had expanded, as it might well have done, the doctrine of discrimination against persons or localities to permit state-wide orders protecting all interstate shippers against discrimination without reference to revenues. At the same time, there remains an opportunity for intrastate shippers or a state commission to show to the Interstate Commerce Commission's satisfaction the existence of specific factors favoring intrastate traffic in general that should not wisely be ignored. It is the solution that seems best designed to achieve the purposes of the Act without interposing insurmountable obstacles to the effective regulation of the national transportation system, the responsibility for which rests, after all, predominantly with the Interstate Commerce Commission.