The Federal Trade Commission petitions for enforcement of its order issued on December 27, 1957, directing the respondent, Standard Motor Products, Inc. (Standard), to cease and desist from charging different net prices to purchasers who compete in the resale of its products, in violation of the Robinson-Patman Act, 49 Stat. 1526 (1936), 15 U.S.C. 13.

The Commission's petition raises two important questions: (1) whether this Court has jurisdiction to enforce cease and desist orders issued by the Commission under the Clayton Act prior to the passage of the Clayton Finality Act, 73 Stat. 243 (1959); and (2) whether the Commission correctly rejected Standard's attempted cost justification of its volume rebate system. We hold that we have jurisdiction to enforce pre-1959 orders, but we hold that in dealing with the important issue of the use of average costs of volume classes of purchasers the Commission failed to articulate criteria reconciling the objectives of the cost justification proviso with those of the Robinson-Patman Act as a whole. We therefore deny enforcement of the order.

The Commission's order, 54 F.T.C. 814 (1957), based upon findings that respondent's volume rebates injured competition among purchasers and were not justified as 'made in good faith to meet an equally low price of a competitor,' 49 Stat. 1526 (1936), 15 U.S.C. 13(b), was affirmed by this Court, 265 F.2d 674 (2 Cir. 1959), and the Supreme Court denied certiorari. 361 U.S. 826, 80 S. Ct. 73, 4 L. Ed. 2d 69 (1959).

The Clayton Act as it stood when the Commission's order was affirmed by this Court in 1959 provided no direct sanction for violation of an order so affirmed. The Commission was required to petition the court of appeals for a decree enforcing the order, which would issue if the court found that the order had been or was about to be violated. Ruberoid Co. v. FTC, 191 F.2d 294 (2 Cir. 1951) (per curiam), aff'd, 343 U.S. 470, 477-480, 72 S. Ct. 800, 96 L. Ed. 1081 (1952). The respondent might then be held in contempt if the enforcement decree was violated. E.g., In re Whitney & Co., 273 F.2d 211 (9 Cir. 1959). This Court has held that if the Commission before petitioning for enforcement of its order conducted a hearing and found that the order had been violated, the Court could treat its findings as though it had been appointed as a master to determine whether violations had occurred. FTC v. Standard Brands, Inc., 189 F.2d 510 (2 Cir. 1951); cf. FTC v. Washington Fish & Oyster Co., Inc., 271 F.2d 39 (9 Cir. 1959).

The Commission accordingly directed Standard to file a report of its compliance with the order which this Court had affirmed. In October 1961 Standard introduced new rebate schedules, which for the first time it sought to justify as making 'only due allowance for differences in the cost of manufacture, sale, or delivery resulting from the different methods or quantities in which (its) commodities are * * * sold or delivered.' 49 Stat. 1526 (1936), 15 U.S.C. 13(a). Although Standard had been informed in June 1961 that the Commission's Division of Accounting has accepted Standard's 1058 cost study on which its new rebate schedules were based, the Commission in March 1962 rejected its compliance report. The Commission directed a compliance hearing, at which Standard presented a 1962 cost study. On the record of the hearing, which was certified to it without any recommendations, the Commission held that Standard had failed to show that its rebates were cost justified, and that it had therefore violated the 1957 order.

Standard contends that this Court lacks jurisdiction to enforce the Commission's order because of the passage of the Clayton Finality Act, 73 Stat. 243 (1959), which amended section 11 of the Clayton Act, 15 U.S.C. 21. We do not agree. The Clayton Finality Act amended the third and fourth paragraphs of section 11, providing for review and enforcement of Commission orders, 'to read as follows,' and set forth an expedited procedure under which an order becomes 'final' upon affirmance by a court of appeals (subject to Supreme Court review) or, where no review has been sought, automatically upon expiration of the sixty days allowed to seek review by a court of appeals. A showing that the respondent has violated the order is not necessary. Violation of a final order draws a civil penalty of not more than $5,000 for each violation, or for each day of a continuing violation.

Section 2 of the Clayton Finality Act expressly preserved old procedure as to certain pending 'proceedings':

'The amendments made by section 1 shall have no application to any proceeding initiated before the date of enactment of this Act under the third or fourth paragraph of section 11 of the (Clayton act). Each such proceeding shall be governed by the provisions of such section as they existed on the day preceding the date of enactment of this Act.' The original third paragraph of section 11 provided for enforcement of Commission orders by a court of appeals, and the fourth for review of orders at the instance of the respondent. 64 Stat. 1127 (1950), as amended.

Immediately after the passage of the Clayton Finality Act, the Commission contended that orders it had previously issued would become final within sixty days, as though they had been issued on the day the Act was passed. The District of Columbia Circuit held, however, that the new procedure applied only to orders issued after the passage of the Act. Sperry Rand Corp. v. FTC, 110 U.S.App.D.C. 1, 288 F.2d 403 (1961). It noted that the Wheeler-Lea Act, 52 Stat. 111 (1938), as amended, 15 U.S.C. 45, which gave orders under the Federal Trade Commission Act the expedited finality which the Clayton Finality Act extended to the Commission's orders under the Clayton Act, had explicitly provided that outstanding orders should become final under the new procedure. The Supreme Court approved the Sperry Rand holding in FTC v. Henry Broch & Co., 368 U.S. 360, 365 n. 5, 82 S. Ct. 431, 7 L. Ed. 2d 353 (1962), and in response to Broch's challenge to the breadth of the order affirmed under the old procedure observed that it could be restricted if and when the commission petitioned for enforcement.

Despite the Broch decision, the Ninth Circuit recently held in FTC v. Jantzen, Inc., 356 F.2d 253 (9 Cir.), cert. granted, 385 U.S. 810, 87 S. Ct. 76, 16 L. Ed. 2d 52 (1966), that the Clayton Finality Act worked an express repeal of the jurisdiction of the courts of appeals to affirm or enforce orders outstanding when the Clayton Finality Act was passed, except as to pending 'proceedings' preserved by section 2. It distinguished Broch on the ground that the respondent in that case had petitioned a court of appeals for review under the old fourth paragraph of section 11 before the passage of the Clayton Finality Act, so that a petition for enforcement would be part of a 'proceeding' saved by section 2. See 356 F.2d at 259. The Ninth Circuit's conclusion that enforcement of the order affirmed in Broch would be authorized by section 2 is clearly correct, since otherwise Broch's petition for review, expressly saved by section 2, would have had to be dismissed as moot. Cf., e.g., Muskrat v. United States, 219 U.S. 346, 360-363, 31 S. Ct. 250, 55 L. Ed. 246 (1911).1 This conclusion is fatal to Standard's contention that this Court lacks jurisdiction to enforce the Commission's order against it, since Standard, like Broch, petitioned for review under the old fourth paragraph of section 11 before the Clayton Finality Act was passed.

We do not rest our holding that we have jurisdiction upon the Ninth Circuit's narrow reading of Broch, however, because we do not agree with the decision in Jantzen. We believe that the old procedure remains available for enforcement of all orders issued before the passage of the Clayton Finality Act. There is no indication in the legislative history of the Act that, congree intended to relegate any class of outstanding orders to the limbo of unenforceability. Under the Jantzen ruling, the stated purpose of the Act, expedition of judicial review and enforcement of Commission orders under the Clayton Act, would be frustrated as to some 400 outstanding orders.

The Ninth Circuit suggested in Jantzen that Congress might well have chosen to abate pre-1959 orders because 'all that the Commission has to do where it finds a violation of the Clayton Act that is also a violation of an old cease and desist order is to enter a new cease and desist order', which will become final and enforceable after a single opportunity for review by a court of appeals. 356 F.2d at 260. But as comentators on Jantzen have observed,2 this conclusion disregards two important factors. First, when the Commission petitions for enforcement under the old procedure, a respondent cannot raise defenses available to him when the order was originally issued, FTC v. Ruberoid Co., 343 U.S. 470, 476-477, 72 S. Ct. 800, 96 L. Ed. 1081 (1952), and may not be able to deny that his price differentials injure competition. See Jaffe, The Judicial Enforcement of Administrative Orders,76 Harv.L.Rev. 865, 897-898 (1963). It is not clear whether he would be similarly barred if a new cease-and-desist order were sought, based upon alleged new violations. Compare, e.g., Commissioner of Internal Revenue v. Sunnen, 333 U.S. 591, 599-603, 68 S. Ct. 715, 92 L. Ed. 898 (1948); FTC v. Raladam Co., 316 U.S. 149, 62 S. Ct. 966, 86 L. Ed. 1336 (1942). See generally 2 Davis, Administrative Law Treatise 18.02-.04 (1958). Second, after the Commission has found that a respondent has violated the Clayton Act, it may sometimes order him to refrain from acts which in themselves would not violate the Act. E.g., FTC v. National Lead Co., 352 U.S. 419, 430, 77 S. Ct. 502, 1 L. Ed. 2d 438 (1957). Thus the Commission would lose important advantages if it had to seek a new order instead of obtaining enforcement of an existing order.

The Ninth Circuit relied in its decision upon the omission of the old procedure when section 11 of the Clayton Act was amended 'to read as follows,' and its express retention for certain limited purposes by section 2 of the Clayton Finality Act. But the rule that amendment 'to read as follows' repeals everything omitted, and the maxim expressio unius set exclusio alterius, are aids in the interpretation of statutes, not syllogisms. They must yield where their application would be inconsistent with the purposes of the statute. E.g., SEC v. C. M. Joiner Leasing Corp., 320 U.S. 344, 350-352, 64 S. Ct. 120, 88 L. Ed. 88 (1943); 1 Sutherland, Statutes and Statutory Construction 2017 (3d ed. Horack 1943); 2 id. 4917. As we have stated, we believe that the Jantzen decision is inconsistent with the purpose of the Clayton Fnality Act.3 For these reasons we conclude that the Commission could petition for enforcement of its 1957 order.

We turn now to whether the order should be enforced in light of the new volume rebates which Standard made effective in 1961 and 1964.

Standard sells two lines of automotive replacement parts, the Standard line of electrical system parts and the Hygrade line of carburetor and speedometer parts. On each line it grants retroactive annual rebates, based on volume of purchases for resale to dealers,4 the rebate percentage rising from one volume class to another. For example, the volume classes and rebate percentages for the Standard line (effective January 1, 1964) are:

 Volume Class    Rebate
---------------  ------
$     0-- 2,999     4%
  3,000-- 5,999    10
  6,000-- 9,999    13
 10,000--24,999    15
 25,000--over      17

Standard's cost justification studies undertook to show that these differing rebate percentages were justified by differences in four classes of selling expenses-- direct selling, catalog, branch were-house, and administrative.

The Commission's basic ground for rejecting Standard's attempted cost justification was the frequent deviation of selling costs attributable to individual purchasers in a volume class from the average cost for that class. While it did not 'reach the question of whether annual volume rebate allowance programs are per se unacceptable,' the Commission held that 'before assigning his customers to a particular rebate bracket, the seller should carefully measure their buying characteristics to be certain that those to be bracketed in all probability will have like cost percentages.' The Commission stated at another point that Standard 'must at least demonstrate that a significant majority of those customers relegated to a particular volume group most likely had costs supporting their inclusion in that group.' In most volume classes in the Standard and Hygrade lines, however, the majority of purchasers had costs closer to the average cost of the class above or below, as the following table illustrates for the Standard line:

                      Percentage
                      With Costs
 Volume    Average     Closer to
  Class     Cost    Another Average
---------  -------  ---------------
$     0--
  2,999    29.51%        29.9%
  3,000--
  5,999    16.11%        54.2%
  6,000--
  9,999    12.89%        70.1%
 10,000--
 24,999     9.41%        67.2%
 25,000--
 39,999     7.58%        70.6%
 40,000--
   over     7.21%        37.5%

The Commission emphasized that it was 'not holding that a seller with a large number of purchasers must individually cost them out * * *.' It thus recognized, as the Supreme Court has stated, that individual cost justification is not required, because it would render uneconomical and impractical even the granting of cost justified price differentials, and 'it seems hardly necessary to say that such a result is at war with Congress' language and purpose.' United States v. Borden Co., 370 U.S. 460, 468, 82 S. Ct. 1309, 1314, 8 L. Ed. 2d 627 (1962); Reid v. Harper & Bros., 235 F.2d 420 (2 Cir.), cert. denied, 352 U.S. 952, 77 S. Ct. 326, 1 L. Ed. 2d 242 (1956); American Can Co. v. Russellville Canning Co., 191 F.2d 38, 59 (8 Cir. 1951). Standard contends, however, that the practical effect of the Commission's rejection of its use of volume classes is to force it to cost justify its rebate to each individual purchaser.

The Commission has not suggested that any administrable alternative means of classifying customers is available to Standard.5 Nor has it indicated the conditions under which it might accept volume rebate schedules for the Standard and Hygrade lines.6 Thus it may well be, as Standard argues, that the Commission has left if no practicable means of cost justification. If so, it is not enough to state, as the Commission does, that 'the obvious result of respondent's discriminatory rebate schedule is that a great number of low-cost customers are burdened with part of the expenses of higher cost purchasers.' This 'economic discrimination'-- charging low-cost and high-cost purchasers the same price, see Rowe, Price Discrimination Under the Robinson-Patman Act 2.2 (1962)-- must be weighed against that which would result from enforced uniform prices, which seem likely to load more of the expense of serving high-cost purchasers upon low-cost purchasers, and which might lead Standard to decide not to sell to small-volume purchasers.7

Moreover, the economic discrimination resulting from Standard's present rebates should have been analyzed more carefully than the Commission has done. For example, Standard adduced evidence that purchasers frequently shift from one volume class to another, or from high to low costs within a volume class, from year to year. Such shifting would seem likely to bring any individual purchaser's average cost over several years closer to the average of his volume class, and thus to lessen the apparent economic discrimination described by the Commission. If so, the Commission should have considered it.8 A careful comparison of the economic discrimination under Standard's present rebates with that under any practicable alternative system of pricing might have led to the conclusion that the present system is the best attainable from this point of view.

Besides analyzing the economic discrimination produced by Standard's rebates, and possible alternative purchaser classifications, the Commission should have attempted to determine whether other sellers in the automobile replacement parts industry, and in other industries, can meet the standards of classification the Commission formulates. If most cannot, there would be serious doubt that the standards adequately respect the purposes of the cost justification proviso.9 We hold, then, that before rejecting Standard's cost justification studies the Commission should have brought its experience and expertise to bear on the problem of defining practicable standards of customer classification for cost justification purposes which reconcile the objectives of the cost justification proviso and of the Robinson-Patman Act as a whole.10

The Commission contends, however, that the standards of customer classification laid down in its report are justified, without any such inquiry as we have held must be made, by the following language of the Supreme Court in United States v. Borden Co., 370 U.S. 460, 468-469, 82 S. Ct. 1309, 1314 (1962):

'But this is not to say that price differentials can be justified on the basis of arbitrary classifications or even classifications which are representative of a numerical majority of the individual members. At some point practical considerations shade into a circumvention of the proviso. A balance is struck by the use of classes for cost justification which are composed of members of such selfsameness as to make the averaging of the cost of dealing with the group a valid and reasonable indicium of the cost of dealing with any specific group mamber. High on the list of 'musts' in the use of the average cost of customer groupings under the proviso of 2(a) is a close resemblance of the individual members of each group on the essential point or points which determine the costs considered.'

We think, however, that this language must be read in light of the distinctive facts in Borden. Borden allowed volume discounts up to 4% on sales of milk to independently owned grocery stores, but granted a flat 8 1/2% discount to two chains.11 Borden attempted to cost justify this large differential by comparing the average cost attributable to the favored chains with that of all other stores, large and small, taken as a single class. Rejecting this attempted justification, immediately after the language set forth above, the Court stated:

'In this regard we do not find the classifications submitted by the appellees to have been shown to be of sufficient homogeneity. Certainly, the cost factors considered were not necessarily encompassed within the manner in which a customer is owned. * * * For instance, the favorable cost comparisons between the chains and the larger independents were for the greater part controlled by the higher average volume of the chain stores in comparison to the average volume of the 80-member class to which these independents were relegated. * * * Such a grouping for cost justification purposes, composed as it is of some independents having volumes comparable to, and in some cases larger than, that of the chain stores, created artificial disparities between the larger independents and the chain stores. It is like averaging one horse and one rabbit. As the Federal Trade Commission said in In the Matter of Champion Spark Plug Co., 50 F.T.C. 30, 43 (1953): 'A cost justification based on the difference between an estimated average cost of selling to one or two large customers and an average cost of selling to all other customers cannot be accepted as a defense to a charge of price discrimination." 370 U.S. at 469-470, 82 S.Ct. at 1314.

The Court was clearly concerned with the use of the average cost of the vast majority of purchasers to cost justify large discounts to the few largest purchasers, a practice which had several times been explicitly condemned. E.g., American Can Co. v. Bruce's Juices, Inc., 187 F.2d 919 (5 Cir.), cert. dismissed, 342 U.S. 875, 72 S. Ct. 165, 96 L. Ed. 657 (1951); Champion Spark Plug Co., 50 F.T.C. 30, 42-43 (1953); International Salt Co., 49 F.T.C. 138, 154-155 (1952). But cf. American Can Co. v. Russellville Canning Co., 191 F.2d 38, 56-60 (8 Cir. 1951). The Government did not question the remainder of Borden's discount schedule, so that the Court was not presented with the very different issue raised by this proceeding of the homogeneity required within each class of a volume discount system with a reasonable number of classes. So far as we can discover, neither courts, Commission, nor commentators have ever analyzed this issue.12 It follows that the Commission cannot rely upon the Court's general language in Borden to relieve it of the obligation to analyze the issue.13

The Commission also rejected several particulars of Standard's cost justification studies. The most important involved the allocation of catalog costs among customers. Standard assumed that the number of catalogs a purchaser received varied in direct proportion to the volume of his purchases, and allocated 'variable' catalog costs accordingly. However, it allocated 'fixed' catalog costs-- the costs of composing the catalogs and readying the presses for the first run-- equally among all customers, on the theory that each required at least one catalog. The Commission held that both types of catalog costs should be allocated in proportion to the number of catalogs received by each purchaser, eliminating all catalog cost differentials between volume classes on Standard's assumption that catalogs distributed varied according to volume, and found that 'with catalog costs removed, respondent's cost study fails in several volume brackets.'

We do not think that this point alone justifies acceptance of the Commission's report. First, while some differentials between volume classes are not fully justified by differences in costs after the Commission's reallocation, the remaining differentials are over-justified by an even greater amount, so that the rebate schedules could be adjusted-- with the end points remaining the same-- so as to be fully cost justified. Second, and more important, we entertain some doubt that the criterion apparently applied by the Commission is rejecting Standard's cost allocation was correct. The Commission apparently did not find that Standard's method of allocation was irrational or not in accord with generally accepted principles; it seems merely to have preferred its own method.14 It might be argued, however, that unless Standard's allocation was demonstrably irrational or out of accord with accounting principles, it should be accepted despite the Commission's preference for another mode of allocation. Compare Int. Rev. Code of 1954, 446(b) (Commissioner may change taxpayer's method of accounting only if it 'does not clearly reflect income'). The inevitably judgmental character of cost allocation might otherwise go far toward making the cost justification defense an illusory one in practice. See Rowe, Price Discrimination Under the Robinson-Patman Act 10.13, at 307 (1962).15

The Commission questioned certain of Standard's other cost allocations as estimates, but apparently none of these caused Standard's studies to fail.

We have carefully considered Standard's procedural objections to the Commission's conduct of the compliance hearing, and with the exception of that discussed in footnote 9 have found them without merit.

Petition for enforcement denied.