In this appeal, petitioners1 contest the August 27, 1984 decision of the Interstate Commerce Commission (Commission or ICC) approving the acquisition by CSX Corporation, the nation’s second largest railroad, of American Commercial Lines, Inc. (ACL), which owns the nation’s largest bargeline, American Commercial Barge Lines (ACBL). Petitioners and intervenors *1181claim that the ICC erred in holding that the transaction violates neither the Panama Canal Act, 49 U.S.C. § 11321, nor section 11344 of the Interstate Commerce Act, 49 U.S.C. § 11344. We find that neither statute prohibits the transaction and, therefore, we uphold the ICC’s interpretation and application of the statutes. We also find that the ICC took the requisite “hard look” at the potential consequences of the merger in its Environmental Assessment of the proposed transaction. We therefore affirm the Commission’s order.
Liquid chemicals, farm products and coal are the three principal commodities served by railroads and bargelines. Over its 27,-000 route miles, CSX — the largest railroad east of the Mississippi River — transports 47.9% of the chemicals, 31.6% of the farm products and 48.3% of the coal moved by all Class I railroads in the East. ACBL, the nation’s largest bargeline, covers 7,500 water route miles and is the only for-hire water carrier which is among the top three for-hire bargelines in each of the three principal commodity markets. In some areas, CSX and ACBL routes parallel each other to a significant degree.
On November 4, 1983, CSX and ACL jointly filed with the ICC an application for CSX to acquire control of ACL and ACBL. In support of the application, the companies asserted that the transaction would achieve substantial public benefits in the form of reduced costs, innovative joint marketing and ratemaking opportunities and other benefits of integrated operations. The companies claim they would be able to offer shippers more efficient, better coordinated, intermodal services, which would expand shippers’ marketing options. In this way, the consolidation would allegedly make CSX and ACL more competitive, thereby enhancing the overall competitiveness of the market.
Various parties participated in the public hearings held before an Administrative Law Judge between February 22 and May 11, 1984. Numerous shippers, states, utilities, water carriers and labor organizations opposed the application on various grounds. Some shippers and states supported the application. The AU made no findings of fact, nor did he prepare a recommended decision. He simply oversaw the hearings. With the benefit of the record generated by the hearings, the ICC heard oral argument on June 21,1984. On August 27,1984, in a lengthy written decision, the ICC approved the acquisition subject to certain oversight and reporting conditions which the ICC imposed as a precaution, to enable it to take corrective action against unforeseen anti-competitive effects.
Essentially, petitioners object to the Commission’s decision on the ground that it eviscerates the Panama Canal Act, 49 U.S.C. § 11321, which prohibits railroad ownership of bargelines except under certain specified conditions. Petitioners do, however, suggest various other grounds for reversal: that the Commission arbitrarily and capriciously concluded that the merger would not result in any reduction of competition prohibited by the Interstate Commerce Act, 49 U.S.C. § 11344; that the Commission denied the public a full and fair hearing by deciding the case without the benefit of an initial AU decision; that the Commission has no authority to impose oversight conditions in a Panama Canal Act case; that the Commission failed to impose protective conditions for the benefit of employees of other railroads; and that the Commission failed to assess the potential environmental effects as thoroughly as is required by the National Environmental Policy Act of 1969, 42 U.S.C. § 4321. We address each of these issues in turn.
Yielding to public concern over the anti-competitive effects of some rail-barge consolidations, Congress passed the Panama Canal Act (the Act) in 1912. The Act generally proscribes railroad ownership of or interest in a water carrier on a water route with which the railroad does or may compete for traffic. The proscription, however, is not absolute. The Commission may authorize the acquisition of such an interest
when the Commission finds that ownership, operation, control, or interest will *1182still allow that water common carrier or vessel to be operated in the public interest advantageously to interstate commerce and that it will still allow competition, without reduction, on the water route in question.
49 U.S.C. § 11321(b).
Only a finding of competition between the acquiring railroad and the acquired bargeline triggers the statutory prohibition and necessitates inquiry into the applicability of the exception. 49 U.S.C. § 11321(a). If the Commission finds no such competition, the Act does not preclude the merger. See generally American Waterways Operators, Inc. v. United States, 386 F.Supp. 799 (D.D.C.1974).
Upon examination, the Commission determined that CSX and ACL do compete, within the meaning of the Act. The Commission then analyzed the merger in terms of the above exception’s two prongs — the public interest prong and the competition prong. Viewing the two prongs as closely related, the Commission found that the public interest prong was satisfied by a showing that the CSX-ACL combination would not be able to flout the public interest by engaging in anticompetitive behavior ultimately allowing it to charge supracom-petitive prices. In the Commission’s view, a finding that that particular harm would not ensue sufficiently ensured the continued operation in the public interest contemplated by the statute. Nonetheless, the Commission also pointed out that anticipated improvements in operating efficiency and quality of service also indicated that the water carrier would continue to be operated in the public interest.
Leaving the public interest prong and turning to competition, the Commission analyzed the competitive nature of the barge industry and concluded that the merger would not be harmful to competition on the waterways. The Commission conceded that the merger could harm individual competitors but, rejecting its own precedent in Illinois Central Railroad Co.-Control-John I. Hay Co., 317 I.C.C. 39 (1962), ruled that harm to competitors is not harm to competition within the meaning of the statute. The Commission also rejected Hay’s holding that the exception required not only a finding that competition would not be reduced among water carriers, but also a finding that competition would not be reduced between water carriers and railroads, including competition between the acquired carrier and the acquiring railroad. The Commission therefore looked only at the effect the merger would have on competition among water carriers, and ignored the effect on rail-barge competition.2 The Commission found that the competitive structure of the barge market is such that the merger does not threaten to harm competition between water carriers.
In petitioners’ view, the Commission has misinterpreted the Act. According to petitioners, the Act reflects Congress’ desire to implement a policy of strict rail-barge separation, subject to only the narrowest exceptions. Petitioners argue that nothing could be further from the narrow exception contemplated by Congress than this merger of giants in their respective modes of transportation. Petitioners also argue that the exception provided for in the Act requires consideration of the effect on rail-barge competition, not just barge-barge competition.
Zealously as petitioners contest the Commission’s interpretation of the Act, they nevertheless also argue that even if that *1183interpretation is correct, the Commission has misapplied the Act. . Petitioners submit that the Commission’s application of the Act is flawed because it is based upon factual determinations regarding the barge industry which petitioners contend are arbitrary and capricious. Specifically, petitioners object to the Commission’s findings that the barge industry is highly competitive and relatively unconcentrated, and that barriers to entry in the industry are not high.
Although substantially intertwined, our review of the Commission’s interpretation of the statute naturally differs somewhat from our review of the Commission’s application of the statute. While an agency’s interpretation of a statute is entitled to deference, “federal courts bear the ultimate responsibility for interpreting federal statutes.” Meade Township v. Andrus, 695 F.2d 1006 (6th Cir.1982). Our task, however, is not to determine whether we think the Commission’s construction is the best construction, but only to determine whether the Commission’s construction was “sufficiently reasonable.” Train v. Natural Resources Defense Council, 421 U.S. 60, 75, 95 S.Ct. 1470, 1479, 43 L.Ed.2d 731 (1975). “To satisfy this standard it is not necessary for a court to find that the agency’s construction was the only reasonable one or even the reading the court would have reached if the question initially had arisen in a judicial proceeding.” Federal Election Commission v. Democratic Senatorial Campaign Committee, 454 U.S. 27, 39, 102 S.Ct. 38, 46, 70 L.Ed.2d 23 (1981).
The characterization of the Panama Canal Act as a strict prohibition against railroad ownership of bargelines runs through all of petitioners’ arguments, and petitioners would have us consider all of their arguments against the imposing background of this characterization. Such characterization, if accurate, would make it difficult to interpret a narrow statutory exception to encompass a merger between companies the size of CSX and ACL. Cur reading of the statute, the legislative history of its initial passage in 1912, and the legislative history of its reenactment in 1940, convinces us, however, that Congress did not intend to mandate absolute rail-barge separation, but, rather, meant to remedy a particular evil which frequently accompanied railroad acquisition of barge-lines.
The “evil” which spurred Congress to action in 1912 was the distasteful propensity of railroads to buy up a bargeline with which they actively competed, lower barge rates on that line sufficiently to drive other barge competitors out of business, then close the acquired bargeline or raise its rates to the level of rail rates, leaving shippers no genuine alternative to rail transport. The large financial resources of the railroads allowed them to weather the short-term loss associated with the reduced barge rates, while forcing the smaller, independent barge companies from the water.
The legislative history is replete with references to this practice. For example, Senator Chamberline remarked:
The people of Portland, Oregon, are practically at the mercy of the same Southern Pacific Co. That company operates a line of railroad from Portland, Oreg., to San Francisco, and at the same time they operate the only line of steam ships between Portland and San Francisco. What is the effect? If they charge a rate which is unfair or too high, an independent line of steamers is put on the river and ocean route. Immediately the railroad-controlled steamers and the railroad itself reduce the rate to a figure which will not afford a reasonable or any compensation to the independent line of steamers. These must then operate at a loss. Then what happens? The independent line of necessity withdraws from competition to avoid bankruptcy, and immediately the railroad company and the railroad-controlled line of steamships raise the rate to more than a remunerative basis. Experience has proven that railroad-controlled steamers can afford to create large deficiencies, which are paid by the railroad company in order to *1184enable the latter to maintain a higher level of rates.
48 Cong.Rec. 10,373 (1912).
Petitioners contend that Congress’ solution to that problem was a strict prohibition against railroad ownership of bargelines. Doubtless some, perhaps many, of the legislators who supported the Act favored strict rail-barge separation. See, e.g., 48 Cong.Rec. 9238 (1912) (“So far as our internal affairs are concerned it would be well to confine our common carriers strictly and exclusively to transportation — the railroads on the land, the boats on the water.”) (Remarks of Senator Townsend). Other supporters of the Act favored the least restrictive legislation necessary to combat the “evil.” Not surprisingly, the legislative history does not conclusively reveal which attitude prevailed. We therefore think it appropriate to heed the oft-cited “canon of construction of the wag who said, when the legislative history is doubtful, go to the statute.” Greenwood v. United States, 350 U.S. 366, 374, 76 S.Ct. 410, 415, 100 L.Ed. 412 (1956).
The Act does not absolutely proscribe railroad ownership of bargelines. Rather, in recognition that the railroads’ motives in such acquisitions were frequently anticom-petitive, the Act promulgates the general rule that railroads shall not own barge-lines. The Act includes the exception, however, and that exception must be read in light of the evil with which the Act concerns itself. The Commission may allow railroad ownership of a bargeline if such ownership “will still allow that water common carrier or vessel to be operated in the public interest advantageously to interstate commerce,” (as opposed to the water carrier eventually being shut down) and such ownership “will still allow competition, without reduction, on the water route in question” (as opposed to reducing competition on the water route by driving other water carriers out of business). The final form of the Act, with its general prohibition and its exception, indicates to us that Congress intended only to end the specific anticompetitive practice which prompted it to act. Congress declined to go further and enact the strict prohibition which some called for and which petitioners now ask us to read into the Act.
This view of the Act finds support both in contemporaneous ICC decisions and in events surrounding the Act’s reenactment in 1940. Among the very first ICC applications of the Act were Lake Line Applications Under Panama Canal Act, 33 I.C.C. 700 (1915), and Ocean Steamship Company of Savannah, 37 I.C.C. 422 (1915). In Lake Line, the Commission considered the applications of several railroads seeking permission for continued ownership of the bargelines which they had owned prior to passage of the Act. The Commission discussed the problem which the statute had addressed, see 33 I.C.C. at 712-13, and, rather than rely on a perceived strict mandate of rail-barge separation, made a searching inquiry into whether the exception should apply. Because the joint operations at issue in Lake Line had attempted “by an artificial rate structure to deprive the public of the natural benefits that would flow from a free use of this waterway” and had caused rates to be “steadily advanced,” 33 I.C.C. at 713-14, the Commission found that the exception did not apply and denied the applications.
Only seven months later, in Ocean Steamship, the Commission applied the exception and granted a railroad permission to continue its ownership of a steamship line with which it competed within the meaning of the Act. The Commission concluded that:
Upon all the facts of record we are of opinion and find that the present operation of the steamship company as a whole is in the interest of the public; that it is of advantage to the convenience and commerce of the people; and that its continued ownership and operation by the Central of Georgia, as at present conducted, will neither exclude, prevent, nor reduce competition on the routes by water under consideration, and that the application should be granted, subject to *1185such further order or orders as may hereafter be entered by the Commission.
37 I.C.C. at 429. From the beginning, then, the exception has been given weight.
In 1940, Congress reenacted the Act. Vigorous debate arose, focusing on a change in wording intended to clarify that the exception could be applied not only to applications for continuing joint operation, but also to applications for new joint operations. The ICC had held precisely this in Southern Pacific Company’s Ownership of Atlantic Steamship Lines, 77 I.C.C. 124 (1923). Again, the debates reveal some sentiment for stricter rail-barge separation, see 86 Cong.Rec. 11537-11544 (1940), but that sentiment did not find its way into the statute as reenacted. As stated by the only other court to consider the question since the 1940 reenactment, “Congress clearly expected rail carriers to be able, with ICC approval, to acquire competing water carriers.” Water Transport Association v. I.C.C., 715 F.2d 581, 590 (D.C.Cir.1983).
Petitioners argue that the ICC’s reading of the Act renders the Act’s prohibition meaningless. If the exception can be read so broadly as to apply to this merger, so the argument goes, it is inconceivable that any other merger could fail to fall within the exception; in effect, the Act no longer prohibits anything. This, say petitioners, amounts to administrative repeal of a statute. While this argument has some merit, it is flawed in that it is subtly wedded to the notion that Congress intended the Act to do more than just address the particular evil discussed above.
Assuming for the moment that it is true, as it may well be, that the Commission’s interpretation of the statute would allow almost any rail-barge merger, the Commission is not necessarily guilty of repealing the Act. Rather, that a merger of this magnitude could, upon analysis, fall within the Act’s exception suggests that the evil which the statute was carefully designed to curtail in 1912 and 1940 is not likely to occur in the transportation industry of 1985. Nor does this amount to administrative or judicial redetermination of a legislatively determined fact. The Act’s flexibility is deliberate. Congress did not determine that all rail-barge mergers worked the same evil. If it had, it would have enacted an absolute prohibition. Rather, Congress contemplated that some mergers might not pose a threat, and those mergers could be allowed. The notion that the day might come when the economics of the transportation industry would preclude many rail-barge mergers from posing a threat to competition is not contrary to any legislative determination. While the findings in the instant case strongly suggest that the Commission believes that day has come, we need not expressly determine that. The Act remains in full force, and the Commission, of course, will continue to analyze acquisitions under the Act on a case-by-case basis.
Petitioners challenge the Commission’s interpretation of the Act in another respect. Petitioners contend that in determining whether the merger “will still allow competition, without reduction, on the water route in question,” the Commission is required to consider the effect of the merger on rail-barge competition, rather than just the effect on barge-barge competition. Under petitioners’ analysis, the elimination of the substantial (or “modest”) competition between CSX and ACL constitutes a reduction in competition under the Act.
The Commission itself previously adhered to that interpretation. In Illinois Central Railroad Co.-Control-John I. Hay Co., 317 I.C.C. 39 (1962), the Commission refused to allow an acquisition because it found that a reduction in competition would follow:
The question is whether the transaction would eliminate, diminish, forestall, hinder, or frustrate competition on the route by water in any manner and to any material degree; as, for example: between Hay and the controlling rail carriers; between Hay and the competing water carriers; or between the competitive water carriers. To decide this question in applicants’ favor it would be necessary to *1186conclude that Hay’s competitive status under the control of the railroads would compare favorably with its present status as an independently operated water carrier. However, such a finding cannot be made on this record and would be improbable in any event in view of the directly competitive nature of the water carrier operations of Hay and the rail operations of the railroad applicants. ... The transaction would reduce competition between Hay and the railroads.
317 I.C.C. at 54 (emphasis added) (citations omitted). In its decision in the instant case, however, the Commission announced that it was changing course:
Finally, in evaluating whether competition will be reduced, we note that preservation of competition in a particular transportation market is not synonymous with the preservation of the position of each competitor.... We must consider whether market forces would preserve the level of competition on the water route generally even though the competitive relationship between the railroad, the acquired water carrier, and other water carriers would be altered. Harm to competitors does not imply harm to competition. Consequently, the holding in ■John I. Hay, to the extent inconsistent with this decision, that a finding of no reduction of competition on the involved water route requires a supporting conclusion that the involved water carrier’s competitive status in relation to other carriers (including the acquiring railroad) would compare favorably with its competitive status as an independent carrier, is rejected and will no longer be followed.
Commission Decision at 19.
Administrative agencies are not bound by their own prior construction of a statute. They are free to reject prior constructions which have not been endorsed by the courts. See NLRB v. Local Union No. 103, International Association of Bridge, Structural & Ornamental Iron Workers, 434 U.S. 335, 351, 98 S.Ct. 651, 660, 54 L.Ed.2d 586 (1978) (“An administrative agency is not disqualified from changing its mind; and when it does, the courts still sit in review of the administrative decision and should not approach the statutory construction issue de novo and without regard to the administrative understanding of the statutes.”) We therefore review the Commission’s construction of the statute without regard to the shift it represents from the construction in Hay.
Petitioners point out that the Act’s prohibition is only triggered by a finding that the acquiring railroad and the acquired bargeline compete. They would argue that it is not reasonable to focus on rail-barge competition in triggering the prohibition, and then to ignore rail-barge competition in construing the exception. We disagree. For one thing, the language of the Act reads, “will still allow competition, without reduction, on the water route in question.” Barges, not trains, operate on water routes. Further, the Commission’s construction is faithful to the Act’s purpose. To prevent the disreputable practice that prompted the Act, it was sensible to use rail-barge competition as a triggering mechanism. No railroad would be tempted to eliminate barge service against which the railroad did not compete for traffic.
Thus triggered, however, no apparent necessity exists for continuing to focus on rail-barge competition. Congress intended to prevent railroads from running their barge competitors ashore. A reduction in competition, even a significant one, between a railroad and a bargeline does not automatically portend a reduction in competition among the various bargelines. If the Commission concludes that no reduction in competition will occur among bargelines, it is irrelevant that some reduction in rail-barge competition may ensue. We therefore hold that it was reasonable for the Commission to construe the competition prong of the Act’s exception as requiring only an analysis of the harm to competition between bargelines.
Even allowing that the Commission was correct to look only at competition between bargelines, petitioners argue that *1187the Commission employed a “destruction of the industry” standard, racher than a mere reduction in competition standard. Petitioners contend that the Commission would not find any reduction in competition unless it determined that all bargelines would be put out of business. Again, petitioners misapprehend the objective of the Act. The Act was not intended to protect barge-lines from the threat of having to compete with a rail-backed bargeline. Had that been the goal, Congress would have enacted a flat prohibition on railroad ownership of barges.
Rather, Congress intended the Act to protect the shipping public from the artificially high rates a railroad could charge after eliminating barge competition. See Steamer Lines on the Chesapeake Bay, 35 I.C.C. 692, 696 (1915) (legislative purpose to insure that “under railroad ownership the lines render service as good as they would render if independently owned and operated, and that railroad ownership does not deprive the public of substantia?, benefits of competition, either in service or in rates”); Lake Line Applications under Panama Canal Act, 33 I.C.C. 700, 712 (1915) (legislative purpose to insure that “[t]he rates charged fluctuate according to economic principles, and the shipper enjoys invariably, as a result, lower charges for the transportation routed over such waterways and thereby reaps a return from the ‘nation’s highway’ ”). In the instant case, the Commission construed the Act to require an actual reduction in competition as opposed to simply harm to particular barge-lines due to intensified competition. As long as the Commission finds that competition will not be so intensified as to drive other bargelines completely out of the market, the interests of the shipping public are protected. We therefore hold that in this respect as well the Commission reasonably construed the Act.
We now turn from petitioners' objections regarding the Commission’s interpretation of the Act to petitioners’ objections regarding the Commission’s application of the Act. The factual determinations which an agency makes in the course of applying a statute, while subject to “searching and careful” review of the whole record, including “the body of evidence which opposes the [agency’s] decision,” Union Carbide Corp. v. NLRB, 714 F.2d 657 (6th Cir.1983), can be rejected only if we find no substantial evidence supporting the conclusion or if we find the conclusion to be so out of touch with the reality revealed by the whole record as to be arbitrary and capricious. See 5 U.S.C. § 706(2).
Petitioners contend that the Commission erred in concluding that both prongs of the Act’s exception were satisfied by the facts in the present case. Petitioners object to the Commission’s treatment of the public interest prong, claiming that the Commission reduced it to mere surplusage by failing to make a specific finding that the merger is essential to the public good. The Act requires no such finding, however. The Commission must find that the acquisition “will still allow that water common carrier or vessel to be operated in the public interest advantageously to interstate commerce.” This is merely a requirement that the public interest not be adversely affected. Increased benefits to the public are not required.
The Commission found that there would be no damage to the public interest, since ACL would continue to operate as an active competitor on the waterways, rather than be shut down by CSX. This finding sufficiently satisfies the public interest prong. Ample evidence supports the finding, and we cannot say that it is arbitrary or capricious.
We now come to the most vigorously debated aspect of the Commission’s decision. In ruling that a CSX-ACL merger “will still allow competition, without reduction, on the water route in question,” the Commission necessarily probed the current nature of the water transportation industry. The Commission was forced to determine facts with respect to various hotly contested questions. The voluminous record generated below contains evidence *1188upon which the Commission might have based conclusions more favorable to petitioners. But other evidence supports the Commission's findings.
It is with respect to these questions, questions intimately entwined with the nature of the industry, that we owe the Commission the greatest deference.
Resolving these considerations is a complex task which requires extensive facilities, expert judgment and considerable knowledge of the transportation industry. Congress left that task to the Commission “to the end that the wisdom and experience of that Commission may be used not only in connection with this form of transportation, but in its coordination of all other forms.” 79 Cong.Rec. 12207.... If the Commission did not exceed the statutory limits within which Congress confined its discretion and its findings are adequate and supported by evidence, it is not our function to upset its order. '
McLean Trucking Co. v. United States, 321 U.S. 67, 87-88, 64 S.Ct. 370, 380-381, 88 L.Ed. 544 (1943). See also Cleveland Electric Illuminating Co. v. I.C.C., 685 F.2d 170, 173 (6th Cir.1982).
The first question is whether, in its competitive analysis of the barge industry, the Commission erred in its market definition. Applying the Department of Justice Guidelines it found the product to be water carrier transportation. It found that although rail transportation afforded some competition to water carriage, that competition was limited because barge rates and costs were so much lower than rail rates and costs and gave only limited consideration to competition from rail and other modes of transportation and only to the extent that they effectively disciplined barge rates on the routes in question. Thus the Commission, rather than distinguishing the three main commodity markets-liquid chemicals, farms products, and coal-initially lumped them all into a single market. Defining the market this way has obvious consequences. The barge industry as a whole appears far less concentrated than, for example, that segment of the barge industry devoted solely to coal transportation. The Commission’s broad market definition reflected an industry in which the leader hauled slightly over five percent of all barge traffic, and the top twelve companies hauled only twenty percent of all traffic. Petitioners point out if the barge transportation of coal is treated as a separate market, three bargelines haul fifty percent of all coal moving on the Ohio River System, which is where the vast majority of coal moves, and the top six barge-lines account for seventy percent of that coal movement. Petitioners also object to including private barge fleets in the relevant markets.
The Commission found that barge equipment was sufficiently interchangeable to permit carriers to switch from one commodity to another and that therefore water carriers specializing in the transport of one commodity can and do switch to other commodities when to their advantage. It also found that the private carriers were competition in the market.
With respect to a geographic market, the Commission included the Mississippi River System and the Gulf International Waterway since equipment was generally interchangeable, and there was no physical barrier to movement throughout those systems. The selection of the relevant market seems to us to be the very type of “complex task which requires ... expert judgment and considerable knowledge of the transportation industry” where the reviewing court should defer to the Commission.
Intervenors challenge the Commission’s findings with respect to interchangeability of barge equipment. They argue that some barges are specialized; those for coal hauling, for example, sometimes having greater inner bottom steel thickness so they will not wear out as soon. The Commission found this to be a carrier’s operational choice but not one that significantly affected the interchangeability. It reached a similar conclusion with respect to problems of covers interfering with the unload*1189ing of coal from barges. The evidence that many barges are used interchangeably even with these differences is substantial evidence to support the Commission’s finding.
Intervenors also argue that the Commission’s finding that there is ease of interchangeability of a coal barge to a grain barge by the addition of a cover is arbitrary since a cover for a typical $225,000 barge costs approximately $50,000, or almost twenty-two percent of the cost of the barge. The $50,000 cost is a substantial sum.
The significance of this sum, however, as an obstacle to transferring barges from coal hauling to grain hauling depends upon numerous factors, such as the profit margin, capital costs, long-term contracts, shipper backing of capital needs, etc. All of these matters are within the special competence of the Commission.
Further, even if we were to hold that the Commission’s finding of interchangeability was arbitrary and capricious, the remedy would be remand for analyses of separate commodity markets. Even though the Commission thought such analysis was unnecessary, the Commission did perform the analysis noting that “[bjecause of the importance of coal and agricultural commodities in the traffic bases of the applicants, we will look at those movements separately and determine whether the transaction would have any adverse impacts on the transportation of those commodities.” Commission Decision at 25.
The Commission analyzed competition for coal hauls and grain hauls in depth. See Commission Decision at 49-58. We see no need to repeat the Commission's analysis. The data employed all find support in the record. Petitioners’ objection amounts to simple disagreement with the Commission on a matter of judgment. Petitioners contend that CSX will channel all the coal shipments originating with it to ACBL and away from other bargelines. Petitioners fear that ultimately they would experience higher unit costs due to lower traffic density, driving them out of business and leaving CSX-ACL to practice monopoly pricing.
The Commission envisioned the ultimate consequences differently. Even looking at a worst-case scenario, where most CSX traffic would eventually be diverted to ACBL from other bargelines, the Commission still concluded that although this would mean diminished traffic for some of ACBL’s competitors, those competitors could still continue to maintain viable operations. The Commission conceded that “any carriers which experience loss of CSX-originated traffic would have to adjust their fleet sizes and operating schedules to the traffic remaining.” Commission Decision at 52. This, however, represents only harm to competitors, not harm to competition, and as the Commission concluded,
we do not think that the assumed “worst case” traffic diversion would harm barge lines competing with ACBL beyond the loss of revenue attributable directly to CSX-originated traffic and associated backhauls. Since these carriers would remain viable competitors, ACBL would face no fewer competitive constraints in its pricing of coal movements after the consolidation than it faces today.
Commission Decision at 52-53.
Petitioners urge upon us that harm to competitors would, in this case, translate into harm to competition by creating an atmosphere in which competitors could not survive. But this is precisely the type of judgment call on which we must defer to the Commission’s expertise in the transportation industry, so long as the judgment is a reasonable one. If competitors will survive, then competition, as contemplated by the Act, will continue unabated. That some competitors may suffer a reduction in traffic does not mean that the competition which benefits shippers will be diminished to any degree. We therefore must decide whether the Commission’s determination that competitors will survive was reasonable.
Petitioners suggest two possible ways in which a CSX-ACL combination could eliminate bargeline competitors: predatory pric*1190ing, which prompted passage of the Act; and “rate scissors,” a sophisticated form of predatory pricing by which a rail-barge combination lowers barge rates and recoups the loss by raising rail rates. The Commission found that neither practice could be successfully implemented by CSX-ACL.
Predatory pricing in these situations was a serious threat in 1912, when a well-financed railroad could absorb a short-term loss while driving less well-financed competitors off the water. The Commission found that predatory pricing would not succeed today, since the barge industry is now “comprised of a large number of firms, many owned or controlled by large and well-financed corporate parents.” These large corporate parents, such as United States Steel and Archer Daniels Midland, are capable of absorbing some losses themselves without throwing in the towel. Further, the Commission reasoned that if predatory pricing could potentially succeed, one of these large corporate parents might already have attempted to eliminate their competitors. Because this has not occurred, the Commission inferred that market forces prevent it.
In the “rate scissors” scenario, a railroad theoretically may engage in predatory pricing without having to absorb any loss; it could, therefore, continue the practice long enough to discourage even the most well-financed competitor. “Rate scissors” is simply a scheme whereby the company sets its barge rates below what its competitors can bear, and makes up the difference in the rail rates it charges customers who have no alternative rail service.
The Commission dismissed the “rate scissors” threat as well, noting that many rail rates are constrained by competition or regulation. If the CSX rail rates rose, many shippers could turn to alternate non-rail options, such as trucks. There was evidence in the record suggesting that the rate scissors approach could actually cause CSX-ACL to lose business. Further, the Commission concluded that CSX-ACL would be likely to employ competing water carriers themselves, in instances where an ACL carrier did not provide the most efficient movement. There was evidence in the record that ACL and other water carriers in fact already do contract outside their own companies when more efficient services are available.
The above factors alone indicate that the Commission’s conclusion that neither predatory pricing nor rate scissors pose a genuine threat to competition is based on a reasoned appraisal of the evidence of record. But the Commission made additional findings that reinforce the conclusion that these threats are illusory. Most significantly, the Commission determined that there are no significant barriers to entry into the barge industry. This ease of entry would tend to discourage predatory practices, since any subsequent attempt to charge supracompetitive rates would likely be met by an influx of competitors undercutting that rate.
Petitioners object that the finding on ease of entry ignores overwhelming evidence to the contrary. They point to the testimony of several large bargeline executives who attempted to break into the market for coal hauls and incurred substantial losses before giving up. This, however, only shows that it is difficult to enter the market at present, when rates and service are presumably optimal, due to competition. The Commission only made the point that because there are no significant regulatory barriers to entry, and because economies of scale may be realized at a relatively low level, a bargeline may enter the market at a reasonable cost, and would be likely to do so should the prevailing rates become artificially high.
We reiterate that the record contains conflicting evidence on all of these points. But, again, our review is limited. When evidence exists to support more than one conclusion, it is best to leave the determination with the Commission, in deference to its familiarity with the industry.
We conclude that the Commission’s interpretation of the Panama Canal Act was *1191reasonable, and its application of the Act’s exception was based on a reasoned and justifiable view of the evidence of record. We affirm the Commission’s decision that the CSX-ACL merger does not violate the Act.
49 U.S.C. §§ 11343-11344 also apply to this transaction. Section 11343 provides that various transactions, including “consolidation or merger of the properties or franchises of at least 2 carriers into one corporation for the ownership, management, and operation of the previously separately owned properties,” may take place only with the Commission’s approval and authorization. Section 11344(d) provides that, except in the case of the merger of two class I railroads, the Commission shall approve the transaction unless
(1) as a result of the transaction, there is likely to be substantial lessening of competition, creation of a monopoly, or restraint of trade in freight surface transportation in any region of the United States; and
(2) the anticompetitive effects of the transaction outweigh the public interest in meeting significant transportation needs.
The inquiry under section 11344 is broader than the previous inquiry, in that the Commission must consider the effect on competition in the industry generally, rather than just among bargelines. The Commission did this, and concluded that the transaction will have no adverse effects on either horizontal or vertical competition in the industry. The Commission’s analysis is thorough and rooted in the record. See Commission Decision at 45-49. It is not our function to reconsider the issue de novo as petitioners would have us do.
Intervenor National Coal Association argues that the Commission failed to expressly weigh the anticompetitive effects against the public interest. This argument overlooks the Commission’s conclusions that no anticompetitive effects would ensue and that the public interest would not be diminished. Given those findings, an explicit balancing of these two factors would be supererogatory. We affirm the Commission’s ruling that the transaction is permissible under 49 U.S.C. § 11344.
We are also presented with a question regarding the Commission’s authority to impose oversight conditions. Petitioners argue that the Panama Canal Act grants no such authority to the Commission. As noted above, however, this transaction also falls within the compass of 49 U.S.C. § 11344. Section 11344(c) expressly provides that the Commission may impose conditions on any transaction approved under that section. Petitioners’ argument to the contrary is without merit.
We also reject petitioners’ contention that the Commission erred in deciding on the contested application without the benefit of an initial or recommended decision by an Administrative Law Judge. While an AU oversaw the hearings which generated the record, the Commission waived the requirement of a separate initial decision by the AU. Petitioners argue that this contravenes the Administrative Procedure Act, 5 U.S.C. § 557(b), which requires an initial decision by the person who presides at the hearings.
Section 557(b)(2), however, provides that, in rulemaking procedures or initial licensing applications, an agency may waive this requirement if the “due and timely execution of its functions imperatively and unavoidably” so require. The Commission waived the requirement on this ground.3
Petitioners object that this case involved neither a rulemaking nor an initial licensing and therefore no waiver is permitted. However, both 49 U.S.C. § 10327(c) and 49 U.S.C. § 11345(f) expressly provide that, in proceedings such as the instant one, the Commission may waive the initial AU decision if such waiver is required “for the *1192timely execution of its functions.” It is true, as petitioners point out, that statutes may not be interpreted to modify the hearing provisions of the Administrative Procedure Act except to the extent that they do so expressly. See 5 U.S.C. § 559. We find, however, that sections 10327(c) and 11345(f) constitute express modifications. It was therefore within the Commission’s discretion to waive an initial decision in attempting to comply with statutorily imposed time limits.4
Petitioner Simmons, the Illinois Legislative Director for the United Transportation Union, takes issue with the Commission’s failure to consider the impact upon and impose protective conditions for employees of non-CSX railroads. He points out that 49 U.S.C. § 11344(b)(1)(B) specifically provides that the Commission shall consider “the interest of carrier employees affected by the proposed transaction.” He cites our decision in Detroit, Toledo & Ironton R. Co. v. United States, 725 F.2d 47 (6th Cir.1984), in support of his argument that we ought to apply this provision in the instant case. See id. at 50 n. 2.
Section 11344(b), however, is expressly restricted to mergers involving at least two class I railroads. Detroit, Toledo & Ironton involved such a merger. The instant case does not. The instant case is subject to the provisions of section 11344(d), covering mergers other than of class I railroads. Section 11344(d) does not contain a requirement similar to section 11344(b)’s requirement to consider the impact on all carrier employees. Section 11344(d) requires only that the Commission consider the public interest. Since section 11344(b) also specifically requires consideration of the public interest, in addition to consideration of the effect on employees, we conclude that Congress did not contemplate that the general notion of “public interest” included the concern about effect on employees.
Simmons argues that I.C.C. v. Railway Labor Association, 315 U.S. 373, 62 S.Ct. 717, 86 L.Ed. 904 (1942), and United States v. Lowden, 308 U.S. 225, 60 S.Ct. 248, 84 L.Ed. 208 (1939), hold that, as a matter of law, the general term “public interest,” in railroad transactions, includes the interest of railroad employees. We find that Railway Labor and Lowden are inapposite. Each case basically held that because the Commission had authority to impose conditions in furtherance of the public interest, it could impose restrictions designed to protect the employees of railroads directly involved in transactions requiring Commission approval. Neither case goes on to hold that "public interest” necessarily includes the interest of employees not directly affected. We hold that the Commission was not required to consider the interests of such employees in the instant transaction.
Simmons also argues that 49 U.S.C. § 11347 requires the Commission to impose protective conditions for the benefit of non-CSX rail employees. Section 11347 provides that in transactions under sections 11344 and 11345, the Commission “shall require the carrier to provide a fair arrangement ... protective of the interest of employees who are affected by the transaction....”
Simmons acknowledges that there is a “split” as to whether this provision applies to employees of carriers not directly involved in the transaction. It is not a very serious split, however. Every court of appeals that has considered the question has concluded that section 11347 does not apply to employees not directly involved in the transaction. See Southern Pacific Transportation Co. v. I.C.C., 736 F.2d 708, 725 *1193(D.C.Cir.1984); Lamoille Valley Railroad v. I.C.C., 711 F.2d 295, 323-34 (D.C.Cir.1983); Brotherhood of Maintenance of Way Employees v. I.C.C., 698 F.2d 315, 316-18 (7th Cir.1983), cert. denied, - U.S. -, 105 S.Ct. 1172, 84 L.Ed.2d 322 (1985). Simmons cites two district court opinions to the contrary, Soo Line Railroad Co. v. United States, 280 F.Supp. 907 (D.Minn.1968), and Railway Labor Executives’ Association v. United States, 216 F.Supp. 101 (E.D.Va.1963).
In Railway Labor, the court interpreted the statute to require protection of employees whom the Commission conceded would be affected even though they were not employed by the railroad involved in the transaction. The court found that although the Chesapeake & Ohio Railway Company was not actually a party to the Commission proceeding, it was “in actuality deeply and unavoidably involved,” and that, due to overlapping interests in a particular train station, many Chesapeake employees “were in reality, though not in contract, as much the employees of Seaboard [the acquiring party before the ICC] as of Chesapeake,” 216 F.Supp. at 103. The court therefore looked only at employees so intimately connected with the transaction as to be “affected” in a meaningful sense. Railway Labor, therefore, does not stand for the broad proposition Simmons urges.
Only Soo Lines stands for the broad proposition that protection must be afforded to all rail employees. See 280 F.Supp. 907, 923-26. We reject the reasoning employed in Soo Lines because it fails to give any meaning to the statute’s use of the word “affected.” Had Congress intended such broad protection, it would not have limited its grant of protection to carrier employees “affected by the transaction.” We hold that section 11347 did not require the Commission to impose protective conditions for the benefit of non-CSX rail employees.
Led by TV A, petitioners contend that the Commission’s environmental review of the proposed merger is inadequate under the National Environmental Protection Act, 42 U.S.C. § 4321-4347 (NEPA). NEPA requires that an agency prepare an Environmental Impact Statement (EIS) before approving any major action that will significantly affect the quality of the human environment. 42 U.S.C. § 4332(2)(C). To determine whether the effect on the environment will be significant enough to warrant an EIS, agencies prepare an environmental assessment (EA). 40 C.F.R. § 1501.4(b)-(c).
An agency decision, based on an EA, that no EIS is required, can be overturned only if it is arbitrary, capricious, or an abuse of discretion. See Sierra Club v. Peterson, 717 F.2d 1409, 1413 (D.C.Cir.1983); Cabinet Mountains Wilderness v. Peterson, 685 F.2d 678, 681 (D.C.Cir.1982). It is not for us to substitute our judgment of the environmental impact for the judgment of the agency, once the agency has adequately studied the issue. It is our role, however, to determine whether the agency has, in fact, adequately studied the issue and taken a “hard look” at the environmental consequences of its decision. See Kleppe v. Sierra Club, 427 U.S. 390, 410 n. 21, 96 S.Ct. 2718, 2730 n. 21, 49 L.Ed.2d 576 (1976); Sierra Club v. Peterson, supra.
In the instant case, the Commission, pursuant to its regulations, directed its Section of Energy and Environment (SEE) to prepare an EA. The EA concluded that the consolidation would have no significant impact warranting preparation of an EIS. The EA purported to examine two sources of potential environmental impact: impacts arising from market extensions of carrier operations and impacts arising from capital improvements projects, such as the three intermodal transfer structures that CSX proposed to construct following approval of the merger.5
*1194The Commission concedes that the scope of its analysis was limited with respect to the contemplated intermodal transfer facilities. The Commission justifies the limited scope on two grounds: first, that the lack of final design and engineering plans made it impossible to conduct an in-depth analysis; and, second, that before the facilities could actually be constructed, in-depth environmental review would be conducted by other authorities.6
Petitioners argue that NEPA requires some degree of speculative forecasting and that, therefore, the lack of definite design information does not justify the Commission’s limited analysis. See, e.g., Scientists’ Institute for Public Information, Inc. v. Atomic Energy Commission, 481 F.2d 1079, 1092 (D.C.Cir.1973). We would agree that the lack of final design plans does not excuse an agency from conducting the most thorough analysis possible of a proposed action. In the instant ease, however, the Commission did not have before it a proposal to construct inter-modal transfer facilities. Rather, the Commission was addressing a proposal that two companies merge. The merger does not inherently require construction of new facilities.
When construction of the new facilities is proposed, the Commission will not be the authority charged with responsibility for deciding on the proposal or preparing any environmental analyses with respect to it. These contemplated facilities are significantly removed from the proposal the Commission was approving. Petitioners argue that CSX and ACL would not desire the merger if they could not build the new facilities. That, however, is a risk that CSX and ACL choose to take. The Commission’s decision in no way permits, nor could it permit, CSX-ACL to construct the contemplated facilities.7
We believe this is precisely the type of situation to which the Supreme Court spoke in Kleppe v. Sierra Club, 427 U.S. 390, 410 n. 20, 96 S.Ct. 2718, 2730 n. 20, 49 L.Ed.2d 576 (1976), where it wrote that NEPA
speaks solely in terms of proposed actions; it does not require an agency to consider the possible environmental impacts of less imminent actions when preparing the impact statement on proposed actions. Should contemplated actions later reach the stage of actual proposals, impact statements on them will take into account the effect of their approach upon the existing environment; and the condition of that environment presumably will reflect earlier proposed actions and their effects.
We do not believe that, in assessing the environmental impact of the merger, the Commission was obligated to assess the *1195impact of contemplated projects which the Commission has no power to approve, which are not an inherent component of the proposed merger, and which will be subject to environmental review should they ever reach the actual proposal stage.
The Commission’s assessment of the environmental impact of traffic increases which may potentially flow from CSX-ACL’s operational changes presents a more difficult problem. 49 C.F.R. § 1105.7 requires merger applicants to file an Environmental Report (ER) answering, among other things, the following question:
Will the proposed action result in (i) a minimum increase in rail traffic of 50 percent or three trains per day on an affected rail line, (ii) an increase in rail yard activity of 20 percent as measured in carload activity or (iii) an increase in motor carrier traffic of either 50 vehicles per day or an increase in truck traffic exceeding 10 percent of the average daily traffic on a given highway segment?
49 C.F.R. § 1105.7(c)(5). If any of those three thresholds will be exceeded, the applicant is required to address, in the ER, the issues of increased air emissions and increased noise levels.
The EA expressly dealt with thresholds (i) and (ii), but did not directly address threshold (iii). When TVA pointed this out in its comments on the EA, the Commission responded, in its Supplemental EA, by stating that, “It is difficult to respond to these arguments except to point out that TVA is simply speculating as to what the actual highway traffic increase might be.” Supplemental EA at 18.
It is unclear whether the Commission’s statement in the EA meant that it was speculative whether CSX-ACL would obtain certain Texas utilities as customers and what percentage of their coal needs it would supply or whether it overlooked CSX-ACL’s comments in which it conceded, that based on certain assumptions “an increase of more than 50 vehicles per day would be experienced for the entire route from each barge unloading site to [certain utilities].” 8 Comments of CSX and ACL at 14.
In light of the opening paragraph of the Supplemental EA which states,
We do not believe that CSX-ACL’s proposals for capital improvement and operational change have been developed to a point where more meaningful environmental analysis of these proposals is possible,
it is reasonable to infer that the Commission took the hard look and concluded that this was not a proposed action. This is confirmed by the Commission’s opinion.
Several commenting parties argue that the environmental consequences of Commission approval of the consolidation warrant preparation of an environmental impact statement (EIS). They contend that NEPA requires that we assess in detail the impacts of the ten potential operational changes and three contemplated capital improvement projects. We disagree. Applicants’ proposals for capital improvement and operational changes have not been developed to a point where more meaningful environmental analysis of these proposals is possible. Applicants’ proposed operating plan considers potential changes and improvements illustrative of what would or might occur as a result of the consolidation. The exact nature or form of the. changes and improvements with respect to environmental analysis is at this point still speculative.
$ $ * $ *
The proposed action before us is our approval of the consolidation. We must consider the environmental impacts of that approval. We are not required, however, to consider the possible environmental impacts of less imminent ac*1196tions. Kleppe v. Sierra Club, 427 U.S. 390, 410 n. 20 [96 S.Ct. 2718, 2730 n. 20, 49 L.Ed.2d 676] (1976). Applicants’ contemplated intermodal plans are less imminent actions. TVA argues that we are the only agency in a position to look at the cumulative impacts of our approval of the consolidation. In respect to such cumulative impacts, however, we are not required to consider actions that are merely contemplated. Hart and Miller, Etc. v. Corps of Engineers, Etc., 505 F.Supp. 732, 752 (D.Md.1980). Furthermore, because applicants have alternate location, for their contemplated facilities on the Tennessee and Ohio Rivers, their plans are not proposed actions. South La. Environmental Council, Inc. v. Sand, 629 F.2d 1005, 1015-16 ([5th Cir.,] 1980).
Commission Decision at 64 & 65. The Commission has taken the requisite hard look.
Accordingly, the decision of the Commission is affirmed.
. Petitioners are eleven water transport companies, the Tennessee Valley Authority, and the Illinois Legislative Director for the United Transportation Union. They are joined by numerous intervenors.
. The Commission also suggested that a finding of no harm to competition could be made if competition between CSX and ACL was so minimal that the merger could not have a substantial anticompetitive impact. Petitioners argue strenuously that finding competition between CSX and ACL to be minimal is arbitrary and capricious. We would agree that such a finding might very well be arbitrary and capricious, but our reading of the ICC decision indicates that, if such a finding was in fact made (the Commission did refer to CSX-ACL competition as "modest"), it did not form the basis for the Commission's conclusions about harm to competition. Rather, the Commission’s analysis of the competition prong relies on its view that only barge competition, and not rail-barge competition, is relevant under the statute.
. "Timely execution” of the Commission’s functions in this instance meant adherence to 49 U.S.C. § 11345(c)(3)’s requirement that the Commission issue a final decision "by the 90th day after the date on which it concludes the evidentiary proceedings.”
. Petitioners’ primary concern over the waiver of an ALJ decision is that they were denied the potential benefit of conclusions the ALJ would have made regarding witness credibility. In this respect, however, it does not appear that petitioners suffered any real harm. Witness credibility would bear only on a limited range of issues in the instant case. The major issues are tied less to the question of whose purported facts to believe than to the question of how to analyze economic data and presage market behavior. Witness credibility plays little or no role in answering this question.
. CSX indicated that, subsequent to the merger, it would seek approval to build new rail-barge transloading facilities at Decatur, Alabama and Philadelphia, Pennsylvania, and to upgrade and add to an existing facility at Louisville, Kentucky.
. The Commission notes that construction at the Louisville site would require some form of approval by the United States Army Corps of Engineers, the Federal Railway Administration, the Kentucky Department of Transportation, the Kentucky Department of Environmental Protection and the Jefferson County Air Pollution Control Office; construction at the Decatur site would require approval by the Army Corps of Engineers, TVA and the Alabama Department of Environmental Management; construction at the Philadelphia site would require approval at least by the Philadelphia Water Department and the Philadelphia Department of Licenses and Permits.
. Petitioners also contend that it is arbitrary for the Commission to view these projects as too speculative for environmental analysis on the one hand, but definite enough to amount to public benefits of the merger on the other hand. We agree that the Commission’s dual approach to the projects is troubling on the surface, but our analysis of the public benefits issue renders this concern moot. The Commission’s discussion of the public benefits of the merger included reference to the efficiencies to be gained from projects such as these. As we said in our earlier discussion of the Panama Canal Act, however, the Commission did not need to go so far as to find increased public benefits, but only that there would be no diminution of public benefits. This conclusion would have been reached by the Commission even without the slightest suggestion that such plants might someday be built. In any event, that the Commission believed construction of the facilities to be probable enough to merit consideration as a peripheral public benefit of the merger did not in itself place the facilities within the scope of issues calling for an environmental analysis by the Commission.
. Presumably, this increase in traffic would result from CSX-ACL’s successful diversion of traffic from competitors which currently ship point to point by rail only, to CSX-ACL, which would make the same point to point shipment by rail-barge-truck, thus adding truck traffic where none had previously existed.