FRIENDLY, Circuit Judge:
Under § 296.11 of the Civil Aeronautics Board’s Economic Regulations, effective October 15, 1948, as amended, air freight forwarders are exempted from the certificate provisions of § 401 of the Federal Aviation Act (other than subdivision (k) (3)), see § 101(3), but must obtain letters of registration. While these are normally issued, under delegated power, by the Bureau of Operating Rights, 14 C.F.R. § 385.13(d) (1967), without a hearing,1 the Board’s policy as we shall later develop in some detail, had been to deny such letters to motor carriers or subsidiaries of motor carriers significantly competing with air carriers.
In January 1966 the Board began a proceeding known as the Motor Carrier-Air Freight Forwarder Investigation, noting that “The fundamental question posed is whether long-haul motor carriers of general commodities should be allowed to participate in air transportation, either in their own right or through the device of acquiring an air freight forwarder.” Various applications for approval of acquisitions of air freight forwarders were consolidated.
These applications had somewhat altered their aspect by the time Examiner [297]*297Ruhlen rendered his initial decision. Two long-haul truckers, D. C. International, Inc. and Navajo Freight Lines, Inc., sought authority to act as air freight forwarders in their own right. Consolidated Freightways, Inc., the nation’s largest long-distance trucker, which had initially sought approval of its acquisition of control of an existing domestic and international air freight forwarder, applied for a letter of registration for its wholly-owned subsidiary, CF Air Freight, Inc., and for the necessary ancillary approvals under §§ 408 and 409. Pacific Inter-mountain Express Company, which also had initially sought approval only of its acquisition of control of an international air freight forwarder, applied both for this and for letters of registration for its subsidiary, P. I. E. Air Freight Forwarding, Inc., and related relief. The applications were opposed by eight scheduled airlines including the six largest domestic operators, by all-cargo air carriers, and by air freight forwarders and their association, most of whom have joined in the instant petition for review. They were opposed also by the Board’s Bureau of Operating Rights.
The Examiner rendered an Initial Decision in April 1967. He began by saying that:
“The immediate issue is whether the four applicants should be authorized to operate as air freight forwarders, but the Board, in instituting this proceeding, stated that the fundamental question is whether long-haul motor carriers of general commodities should be allowed to act as air freight forwarders.”
He considered that the issue must be resolved against the background of “free entry” into air freight forwarding which he regarded as established in the Air Freight Forwarder Case, 9 C.A.B. 473 (1948). Reviewing various contentions of the applicants as to what their position as motor carriers would enable them to contribute to air transportation, he found these claims of special advantages were not made out, save only that “use of the sophisticated accounting, billing, and reporting systems which only a large organization can afford should assist the applicants in providing a convenient service to the public.” Nevertheless he concluded that “the applicants, by providing sophisticated management, personnel, and equipment and devoting their well-financed experienced organization to the generation and transportation of cargo by air, will contribute a substantial benefit to the public.” Turning to the effect on existing air freight forwarders, he thought “the strong and efficient forwarders will suffer only minor losses with the entry of the applicants into competition,” although “the demise of some of the weaker air freight forwarders may be hastened by the competition provided by the applicants.” Believing that “If the applicants are to be successful, they will have to rely on newly generated traffic,” the Examiner dismissed the arguments based on their conflict of interest primarily on the basis that “although in some circumstances for short periods of time an energetic and clever promoter may be able to persuade shippers to use modes of transportation adverse to their economic interests, in the long run, the shipper will choose that form of transportation which best meets his needs.” Finally, following a previous Board ruling that the second proviso to § 408(b) 2 was inapplicable to “indirect air carriers,” Air Freight Forwarder Case, supra, 9 C.A.B. at 503, reaffirmed in Airfreight Forwarder Investigation, 21 C.A.B. 536, 544-45 (1955), he held there was [298]*298no legal barrier to the applications, and proposed that all should be granted for an experimental period of five years.
The Board, over a dissent by Vice Chairman Murphy, followed the Examiner’s recommendations except in a minor respect not here material.3 However, the Board exhibited markedly greater enthusiasm over the new departure than had the Examiner. Apparently dissatisfied with the Examiner’s theory that the conflict of interest arising from authorizing long distance truckers to act as air freight forwarders should be disregarded because it would not be effective in the long run, the Board decided that the double role would lessen conflict rather than increase it. Whereas the Examiner seemed to have thought that generation of new air freight by the truckers would be only of the same sort that any vigorous sales effort would engender, the Board found a particularized source in the “large proportion of the truckers’ present surface traffic * * * comprised of shipments weighing less than 200 pounds,” which they now transport “at a loss or minimal gain,” and whence they “could earn more as forwarders by shipping such packages by air.” It concluded also that authorizing the truckers to act as air freight forwarders “should increase the intermodal carriage of freight by air and truck”; indeed, it regarded its decision “as a real breakthrough in opening up the most hopeful avenue for increasing intermodal transportation of freight by surface and air.” It thought also that “The participation * * * of motor carriers like the applicants may well be necessary to achieve the full promise of air cargo,” emphasizing especially the role they could play in areas outside the ten cities in which the traffic is most developed. For these reasons, as well as those urged by the Examiner, it was “convinced that a new policy towards motor carriers like the applicants deserves a trial.”
I.
The Board’s decision does not measure up to the standards required by § 8(b) of the APA, as recently applied in Northeast Airlines, Inc. v. CAB, 331 F.2d 579 (1 Cir. 1964), see especially Judge Aldrich’s concurring opinion, 331 F.2d at 589. The most serious deficiency lies in the ambiguity whether the Board has established a policy of entry for all truckers who want to act as air freight forwarders or has merely granted the four applications that were before it, and the inadequacy of its consideration of effect on the existing forwarders if the former is the right interpretation as we strongly suspect.
Certainly the Examiner thought, as instanced by our first quotation, that the decision would have general applicability. That also would be the natural inference from such statements by the Board as that “Our purpose in instituting this investigation was to determine whether our traditional restrictions on surface carriers have become outmoded for motor carriers like the applicants,” that it regarded its decision “as a real breakthrough” in stimulating intermodal transportation, and that it was establishing “a new policy towards motor carriers like the applicants.” Indeed the Board’s brief in this court concedes that “its findings and pronouncements also were such as to indicate that it may be expected to grant future applications by other long-haul motor carriers to engage in air freight forwarding activities” — which, as indicated, is normally done by its. staff and without a hearing.
On the other hand, the agency almost completely failed to examine the effect that entry of large numbers of motor carriers will have upon the structure of the air freight forwarding industry. The Board’s opinion does not really consider this question at all — its finding that “motor carriers like the applicants” [299]*299would stimulate the development of air cargo is based entirely upon a review of the operational plans submitted by the four carriers in the instant proceeding and no attempt is made to assess the effect of a general motor carrier invasion. The Examiner’s analysis is only slightly less scanty. At one point the possibility of adverse effect upon the industry is discounted on the ground that “the independent freight forwarders” are in an “ impregnable” competitive position at the present time. But to evidence this conclusion, the Examiner pointed only to the increasing profitability of the top ten forwarders. As we have already noted, other parts of his opinion recognize that “the demise of some of the weaker air freight forwarders may be hastened by the competition provided by the applicants.” If “some” of the existing forwarders will succumb to the challenge of these four motor carriers, the impact of a general policy admitting all long-haul truckers will be much more severe, and the danger is enhanced by the probability, hereafter developed, that the truckers would bend their efforts primarily to forwarding freight now moving by air rather than to diverting freight now moving by truck.
The record shows that the air freight forwarding industry is now very much divided between the haves and have-nots. Of the 120 forwarders the ten largest accounted for 72.8% of total revenue and earned 7.2 millions dollars in pre-tax profits in 1965; the remaining 110 suffered a net operating deficit of 1.1 million dollars. Seventy percent of the domestic forwarders earned less than $25,000 in 1964; one-third operated at a loss. Indeed, even the aggregate figure representing the profitability of the “impregnable” top ten is quite deceiving — of the 7.2 million dollars earned by these forwarders, 5.2 was garnered by Emery Air Freight, which has consistently captured 30% of the market. Taken as a whole the industry would thus seem highly vulnerable to the effect of entry by many well-financed competitors.
As to the number of truckers that might take advantage of the Board’s new policy, the Bureau of Operating Rights presented figures that reveal the dimension of the problem:
Numbers of Carriers in 1964 Compared by Revenue Characteristics (Exhibit BOR-8)
$1,000,000 or More Gross Revenue $200,000 to 1,000,000 Under 200,000 Total
ICC Licensed
Motor Carriers 1,195 2,536 11,748 15,479
CAB Authorized
Air Freight
Forwarders 17 32 56 105
The dangers seem even greater than this chart reveals. There are 50 truckers with gross revenue exceeding $25,-000,000 and 20 with gross revenue of some $40,000,000 or more — a level attained in air freight forwarding only by Emery Air Freight, with 30% of the market, even when Emery’s thriving international forwarding operations are included. Indeed, three of the four instant applicants have revenues larger than even Emery — Consolidated Freightways being three times as large. Only three other air freight forwarders who are members the forwarders’ association, one of them predominantly international, have gross [300]*300revenues of $10,000,000 or more, as against over a hundred truckers with revenues exceeding that amount. In this context, the Examiner’s confidence in the ability of even the stronger independent forwarders to withstand a massive invasion of motor carriers is without substantial evidentiary support.4
The record affords no indication that the agency made any informed determination of the number of motor carriers that would seek to take advantage of its new policy. While the fact that only four have applied could indicate that the others lack interest, this seems more likely to have represented simply a tactical determination that it was best for a small number to try to open the gate. Indeed, competitive reasons might force other motor carriers to follow the leaders whether or not they would otherwise have wished. The Bureau of Operating Rights seemed to suggest this when it said in commenting on the chart we have reproduced: ‘If, for competitive reasons, all of the motor carriers in the over $1,-000,000 revenue category apply for and are granted air freight forwarder authority, the size of the air freight forwarder industry would increase by more than 10 fold.” 5 It is as obvious as anything can be that the effect on the existing air freight forwarders will be very different if the permission here accorded is utilized by all long-haul motor carriers, almost all, many, or only four. And, as the Board properly recognized, the continued health of the independent forwarders is of utmost importance not only to themselves but to the future of air freight. Indeed the argument that motor carrier applicants were not disqualified by their conflicting interests rested in part on the continued ability of the independents to prevent them from fooling all of the people all of the time.
Perhaps concerned by realization that the breath of the Board’s policy pronouncement and the meagreness of its findings as to competitive effect were not in mesh, counsel for the Board, in a footnote to their brief, accused petitioners of having “distorted the Board’s opinion” by indicating “that only casual pro forma consideration would be given to further applications of ‘all members of’ an ‘entire class’ * * We find no distortion; indeed, as already indicated, we would so read the opinion and Board counsel themselves said as much in the earlier passage we have quoted from their brief. Of course, it may be that the footnote is right and the earlier passage wrong; if entry of motor carriers into air freight forwarding is indeed to be an experiment, it would seem much more sensible to limit the number unless — another point on which the decisions shed no light — this would create unfair dislocations within the motor carrier industry. But the case in this aspect comes precisely within Mr. Justice Cardozo’s much-quoted observation, “We must know what a decision means before the duty becomes ours to say whether it is right or wrong.” United States v. Chicago, M. St. P. & P. R. R., 294 U.S. 499, 511, 55 S.Ct. 462, 467, 79 L.Ed. 1023 (1935); see also Secretary of Agriculture of United States v. United States, 347 U.S. 645, 654, 74 S.Ct. 826, 98 L.Ed. 1015 (1954).
[301]*301This failure of analysis of the effect of general entry by motor carriers cannot be excused on the basis of the Board’s twenty year commitment to a policy of “free entry” into freight forwarding in contrast with the contrary view arrived at by the Interstate Commerce Commission and adopted by Congress as to surface forwarding. See 71 Stat. 452 (1957), and 2 U.S.Code Cong. & Adm. News, 85th Cong. 1st Sess. pp. 1640-45 (1957). Free entry by persons willing to risk new capital in air freight forwarding as a prime occupation is one thing; entry as a sideline to much larger investments in other business is something else again. This consideration is peculiarly important when the other business is surface transportation — another point of difficulty to which we now turn.
II.
Until the decision here under review, the Board has granted surface carriers air forwarding authority only after study of the particularized situation convinced it that the carrier’s conflict of interest would not “result in material diversion of traffic from air to surface transportation and deprive the applicants of sufficient incentive to conscientiously promote and develop airfreight forwarding.” Air Freight Forwarder Authority Case, Order E-21056, July 10, 1964 (mimeo, page 4). In the first Air Freight Forwarder Case, supra, 9 C.A.B. at 507-08, the Board granted authority to three Class I motor carrier affiliates but only on a showing that the motor carrier operations were so limited as not to be truly competitive with air transportation ; it denied authority to a subsidiary of a railroad on the grounds that this “would create a serious handicap to independent air freight forwarders” and that the conflict of interest “might prove detrimental to the full development of the air services.”6 In it's second domestic Airfreight Forwarder Investigation, 21 C.A.B. 536, 546 (1955), the Board granted experimental authority to two forwarders controlled by railroads, but once again the decision was made only after the applicants’ particular proposals were studied, and no indication was given that subsequent applications would not be subjected to similar scrutiny. Finally in 1964, the Board granted motor carriers permission to forward household goods only after finding that very little of this merchandise was then being shipped by air and that carriers of household goods did not have substantial investments in long-haul equipment competitive with air transit.7 Compare Telstar Air Freight, Inc., Order E-22429 (issued under delegated authority) (1965). What was done here was not a modest expansion of these exceptions but a total reversal of policy.
[302]*302We fully agree that, as Mr. Justice Fortas recently so well expressed it, an administrative agency, “faced with new developments or in light of reconsideration of the relevant facts and its mandate, may alter its past interpretation and overturn past administrative rulings and practice. * * * [T]his kind of flexibility and adaptability to changing needs and patterns of transportation is an essential part of the office of a regulatory agency. Regulatory agencies do not establish rules of conduct to last forever; they are supposed, within the limits of the law and of fair and prudent administration, to adapt their rules and practices to the Nation’s needs in a volatile, changing economy. They are neither required nor supposed to regulate the present and the future within the inflexible limits of yesterday.” American Trucking Ass’ns v. Atchison, Topeka & S. Fe R.R., 387 U.S. 397, 416, 87 S.Ct. 1608, 1618, 18 L.Ed.2d 585 (1967). But we also concur with the First Circuit that a reversal of agency policy calls for thorough analysis and explication, City of Lawrence, Massachusetts v. CAB, 343 F.2d 583 (1965), such as was performed by the Interstate Commerce Commission in the case eliciting the remarks of the Supreme Court we have quoted.
This is particularly necessary when the previous policy is not wholly agency determined but is a radiation from a direction of Congress. We need not quarrel with the Board’s previously cited decision, sustained by the Seventh Circuit, American Airlines, Inc. v. CAB, 178 F.2d 903, 909-910 (1949), that the second proviso to § 408(b), see fn. 2, does not apply in terms to an “indirect air carrier” since such a carrier could never avail itself of the statutory escape hatch, namely, demonstrating that it can “use aircraft to public advantage in its operation.” Nevertheless, as was held by another court of appeals reviewing the same decision, in an opinion curiously not called to our attention, National Air Freight Forwarding Corp. v. CAB, 90 U.S.App.D.C. 330, 197 F.2d 384 (1952), the Board is nevertheless obliged to consider the policies behind the proviso in regulating entry into forwarding. Congressional efforts to prevent participants in one form of transportation from limiting the development of competing forms have a fifty-five year history, going back to the Panama Canal Act of 1912, 49 U.S.C. § 5(14)-(16), see Lake Line Applications Under Panama Canal Act, 33 ICC 699, 710-14 (1915). As to the importance of this policy in air regulation, we need only refer to the various opinions in American President Lines, Ltd., Petition, 7 C.A.B. 799 (1947), where, after thorough reexamination, a majority of the Board concluded that although the second proviso to § 408(b) did not in terms apply to an application by a surface carrier for a certificate under § 401, respect for the policy determination of Congress required that unless such a carrier could meet the qualification of the second proviso as to using aircraft in its operation, it must show “that disadvantages existing upon the facts of the particular case by reason of its being a surface carrier were avoided or overcome by other considerations of public interest supported by the record.” 7 C.A.B. at 807.8 Later decisions of the Supreme Court in the motor carrier field show that the effect of the second proviso is at least that great. See American Trucking Ass’ns v. United States, 355 U.S. 141, 78 S.Ct. 165, 2 L.Ed.2d 158 (1957); American Trucking Ass’ns v. Frisco Transportation Co., 358 U.S. 133, 141, 79 S.Ct. 170, 3 L.Ed.2d 172 (1958); American Trucking Ass’ns Inc. v. United States, 364 U.S. 1, 80 S.Ct. 1570, 4 L.Ed.2d 1527 (1960).
We find no force in the contention of Board counsel that questions of conflict in interest in air freight forwarding were ruled out by a clause in § 410(c) of Title IV added to the Interstate Commerce Act by the Freight Forwarder Act of 1942, 56 Stat. 291, which we quote in [303]*303the margin.9 Apart from the fact that the clause prohibits the issuance of forwarder permits to carriers themselves, as the Board here proposes to do for D. C. International and Navajo, since “this would create difficulties of administration and possible discriminatory practices,” 87 Cong.Rec. 8427 (1942) (remarks of Representative Wolverton),10 the policies behind the 1942 proviso do not apply to the situation before us. Congress believed that since independent forwarders could consolidate shipments, thereby gaining a profit, it was “fundamentally unfair to deny to a common carrier that has invested its money in transportation facilities the right to use those facilities upon as favorable a basis as any forwarder can use them.” Id. So far as air forwarding is concerned, it would be the airlines, not other common carriers, that could make a similar complaint. A second reason for this provision, also inapplicable here, was that certain railroads had already acquired forwarders whereas others had not and it was desired to avoid discrimination. See U. S. Code & Adm. News, 85th Cong. 1st Sess. p. 1641 (1957); 88 Cong.Rec. 8427 (1942) (remarks of Representative Wolverton). Since Congress was careful to make clear that “the Interstate Commerce Commission shall not have jurisdiction of, and part IV shall not apply to, freight forwarding by the use of air facilities, utilized in connection with air transportation * * H.Rep.No. 1172, supra at p. 7 (1941), the proviso to § 410(c) of the Interstate Commerce Act is of no importance when the concerns that motivated its enactment are found to be irrelevant.11 See also Pan American World Airways, Inc. v. CAB, 380 F.2d 770, 782 (2 Cir. 1967), cert. granted, 390 U.S. 919, 88 S.Ct. 850, 19 L.Ed.2d 978 (1968).
We recognize that there is a substantial difference between allowing a surface carrier to control or become one of the limited number of certificated air carriers and to enter the field of air freight forwarding where a larger number of competing operators can thrive. Nevertheless, the Board must make adequately supported findings either that the suspected conflict of interest does not exist or that important public advantages justify disregarding it. The difficulty in making the former finding here would seem rather formidable in view of the Examiner’s conclusion, accepted by all concerned, that “the rate of growth of air-cargo transportation will depend substantially upon the extent to which it is scientifically promoted * * * ” and that the “largest potential source of air cargo” will only be developed as a result of “in-depth analysis of a shipper’s operating problems” — this to include “determining the comparative cost of warehousing, damage claims, communications, storage, use of money, size of inventories, packing costs, and many other factors.” [304]*304It is thus of little consequence that a motor carrier will readily forward shipments by air when the shipper requests this. The problem arises when the shipper is not already convinced of the merits of air freight, and we find a certain unreality in the spectacle of a long-haul motor carrier urging a shipper that higher air rates are counterbalanced by lower expense for warehousing, damage and pilferage, smaller packing costs, and inventory savings, when he knows that such arguments may spread from shipments he is willing to see go by air to others he decidedly is not. Yet, as the Examiner recognized, the forwarder is a kind of transportation adviser, see Comment, Intermodal Transportation and the Freight Forwarder, 76 Yale L.J. 1360, 1361-62 (1967), and his services are generally used by the smaller unsophisticated shipper. See Snow, Air Freight Forwarding: A Legal and Economic Analysis, 32 J. of Air Law & Com. 485, 489-92 and table 4 (1966). To argue, as the Examiner did, that because the motor carriers cannot prevent certain shipments from moving by air, they will try to promote their competitors’ service in marginal cases seems at war with common sense. His alternative argument, that the motor carriers would be forced to promote air shipment by the efforts of the independent forwarders, presupposes continuing vigor of the independent forwarders and, apart from other objections, the lack of evidentiary support for this premise in the context of a large-scale invasion by motor carriers causes the conclusion to fall with it.
The Board sought to strengthen the case by proffering two additional considerations. In something of a tour de force, it argued that granting air freight forwarding authority to competing motor carriers not merely would present no real problem of conflicting interest but “will in fact reduce whatever conflict of interest may exist.” This was because “if there be a conflict of interest, it inheres in the circumstances where the trucker may have a choice with respect to any shipment between transporting it all the way himself or turning it over to air carriers for part of the journey.” If truckers have an economic incentive to ship by air in these situations, the Board concluded, their “conflicting” interest in the surface long haul will thereby be diminished. This betrays a serious confusion of thought. The interest of a trucker, so long as he sticks to trucking, is to truck to the utmost, save for shipments he will carry at a loss; he can have no conflict of interest even though he may cast a longing eye on other pastures. Conflict of interest arises only when an operator is permitted to engage in competing modes of transportation, with the inevitable consequence that his superior economic stake in one may conflict with his using the other in the best interests of the customer. If a trucker advises a shipper that air freight is uneconomic, even a naive businessman would recognize this to be special pleading. It is only when the trucker becomes an air freight forwarder, ostensibly also interested in air transport, that he is cast in a role creating danger of his being able to effect unfair preservation of a status in which surface carriage is the norm. While the Board might conclude, after studying the proposals of a particular surface carrier, that despite its conflict of interest the applicant will conscientiously promote air transport, it cannot simply wish the conflict out of existence for a whole industry as it has done here.
Perhaps it was some awareness of this fallacy in its general position as to conflict of interest that led the Board to mark out a category of traffic, to wit, packages of less than 200 pounds, where no conflict was thought to exist since the truckers now carry these at a loss or minimal gain and they could earn more as forwarders by shipping such packages by air. However, the point outruns the evidentiary basis for it. The sole exhibit on which this conclusion rested had been presented by the Board’s Bureau of Operating Rights with the caveat that its purpose was “not to attempt to fix the cost of handling shipments in relation to the revenue yields associated with spe[305]*305cific commodities,” but simply to suggest “that different weights and different commodities may have a significantly different impact upon motor carrier costs and revenues.” To that end the Bureau took shipments of metal products, machinery and chemicals, with weight breaks at 100, 200 and 500 pounds, and compared transcontinental motor carrier average costs with Consolidated Freight-ways’ revenue yield for sample shipments over similar mileages. The figures indicated a loss for metal products and chemicals in all three weight-breaks and for machinery below 200 pounds. Some things the figures do not tell are how far transcontinental average costs are appropriate for Consolidated Freightways, whether incremental costs would not be more significant for the use to which the Board put the exhibit, how far the figures for these three categories of traffic are representative, to what degree truckers would prefer to carry the smaller packages at a loss in order to prevent the shipper’s becoming accustomed to air freight, and how far the value of these shipments as compared with their weight and bulk would make them candidates for air freight forwarding so that the truckers might earn more by sending them that way. Not only is the exhibit inadequate on its face for the purpose for which it was used,12 but other exhibits submitted by the Bureau squarely contradict the notion that motor carriers would gladly divert a large number of small shipments to the airlines. An I. C. C. study of a representative group of long haul motor carriers, including all four applicants, found that 45% of the shipments transported by the truckers weighed less than 150 pounds. From this exhibit and others, the Bureau concluded that “motor carriers move the same com-modifies in small weight break categories as the air carriers. These commodities provide a primary source of weight and revenue for the direct air carriers, the air freight forwarders and the motor carriers.” It is exceedingly dangerous business — indeed under some circumstances a denial of the fair hearing required by the APA — for an agency to seize upon a single exhibit in a large record, introduced for a limited and hardly controversial purpose, and make this an important basis for decision without benefit of cross-examination, rebuttal or argument. Compare Ohio Bell Telephone Co. v. Public Utilities Comm’n of Ohio, 301 U.S. 292, 304-306, 57 S.Ct. 724, 81 L.Ed. 1093 (1937).
III.
The Board’s failure to come to terms with the conflict of interest problem also undermines its findings that important public advantages will result from motor carrier entry into air freight forwarding. The agency emphasized that the truckers can be expected to develop air cargo in markets in which this service is presently being underutilized; this was of particular importance with respect to the further conclusion that the authorization would greatly increase intermodal carriage since that consideration faced an initial obstacle in the Examiner’s uncontradicted finding that “most of the air cargo travels between major airline points * * But it is precisely in these anticipated new markets, where there is at present little independent forwarder competition, in which the conflict of interest argument tells most strongly. Large numbers of independent forwarders have been entering the industry in recent years — the number of forwarders increased by 54%- between 1961 and 1965 [306]*306and on December 31, 1965, sixty more forwarders had 98 applications for domestic and international authority pending. If a large number of truckers entered the underdeveloped markets at the present time and proceeded to follow their economic interests by skimming the cream, prospective independents might hesitate long before deciding to enter, cf. F. T. C. v. Procter & Gamble Co., 386 U. S. 568, 579, 87 S.Ct. 1224, 18 L.Ed.2d 303 (1967); Bain, Industrial Organization 239-40 (1959). Here again, the Board did not consider this problem in industry-wide terms, but simply based its conclusion on the applications of the four truckers in the present proceeding.
The Board also concluded that the grant of forwarding authority will induce the motor carriers to develop air cargo traffic beyond the traditional radius of 25 miles or within the city limits, 14 C.F.R. § 222.21 (1967), 49 C.F.R. § 210.40 (1967), to which the other air freight forwarders have been restricted. In its opinion, however, the Board promised to consider expanding these limits in accordance with recent judicial decisions, Law Motor Freight, Inc. v. CAB, 364 F.2d 139 (1 Cir. 1966); National Motor Freight Traffic Ass’n, Inc. v. CAB, 374 F.2d 266 (D.C.Cir. 1966), cert. denied in both cases, 387 U.S. 905, 87 S.Ct. 1687, 18 L.Ed.2d 623 (1967),13 and petitioners argue that the agency’s previous failure to take this step cannot provide it with a reason for permitting the truckers to enter the industry. To some extent this argument neglects that the Interstate Commerce Commission must also approve of such an expansion of service, see 49 U.S.C. § 303(b) (7a) (1964), Air Dispatch, Inc. v. United States, 237 F.Supp. 450 (E.D.Pa.1964), aff’d per curiam 381 U.S. 412, 85 S.Ct. 1576, 14 L.Ed.2d 693 (1965). While that agency’s approach has been criticized as unduly restrictive, Note, Regulation of Air Freight Pickup and Delivery, 76 Yale L.J. 405 (1966), it nevertheless has declared that it would not be bound by the CAB’s decision, although giving weight to this. Motor Transportation of Property Incidental to Air, 95 M.C.C. 71, 87 (1964), aff’d sub nom. Air Dispatch Inc. v. United States, supra; cf. Zantop Air Transport, Inc. v. United States, 272 F.Supp. 265 (E.D.Mich.1967). But while the Board was justified in considering the truckers’ ability to solicit air freight along their regular routes outside the area where air forwarders can truck, once more it has failed sufficiently to explain why, as a general matter, they would be motivated to do so.
A final point of concern, although we should not have reversed on this ground, is the failure of the Examiner and the Board to relate their decision in depth to one matter on which every one was agreed. This was that with the introduction of additional jet-propelled cargo-only aircraft and more mechanization, air freight costs and rates will decrease while service will improve, so that air competition with surface transportation will become even more intense. At first blush this would seem a strange context for the grant of authorizations which, if followed as a general policy, would give motor carriers an important, if not indeed a dominant, role in air freight forwarding. Granted that the increased capacity will [307]*307demand ever greater sales efforts, one would wish to know why this could not be better afforded by independent air freight forwarders, now existing or entering the market as a result of the heightened prospects. We fully recognize that, as Judge Prettyman classically stated in American Airlines, Inc. v. CAB, 192 F.2d 417, 422 (D.C.Cir. 1951), agencies are not required to forecast the future “with mathematical precision.” But the very bite of his fine opinion is that in granting new authorizations, an expert agency must look forward and make the best informed estimate that it can. We find it disappointing that, on an issue of such importance to the instant determination, that function was not more fully performed.
We would not wish to be misunderstood as ruling that even the present record might not be sufficient for the Board to initiate a properly controlled experiment in the authorization of truckers as air freight forwarders, with the limitation on numbers and the reporting and other requirements an experiment would be expected to entail. But that is not at all what the Board did. Reading the opinion as a commitment to grant authorizations to all comers, we find it wanting in the careful investigation, the substantial evidence and the rational explication that are demanded before an expert agency may lawfully embark on a new course apparently so fraught with danger to the industry Congress has confided to its regulation and seemingly so opposed to the general policy Congress has long decreed.
The order is vacated and the cause remanded for further proceedings not inconsistent with this opinion.