The opinion
HILL, Circuit Judge,
dissenting:
I respectfully dissent from the majority’s resolution of the issue of state action immunity. In evaluating the effect of California Retail Liquor Dealers Ass’n v. Midcal Aluminum, Inc., 445 U.S. 97 , 100 S.Ct. 937 , 63 L.Ed.2d 233 (1980), on this case, I am persuaded that it establishes a standard different from that applied by the district court 1 and that the majority has misinterpreted the Mideal standard. I would therefore reverse and remand in part for further consideration.
The central issue in this case — whether the practice of collective rate formulation employed by the rate bureaus is immune from antitrust attack under the state action doctrine — is one beset with difficulties. Of state action immunity it is well said that “[ujnlike many other phantasmagoria of the law, we know it is there, but it is hard to define precisely what it is, and, unlike obscenity, we do not always know it even when we see it.” Kennedy, Of Lawyers, Lightbulbs, and Raisins: An Analysis of the State Action Doctrine Under the Antitrust Laws, 74 Nw.U.L.Rev. 31, 31 (1979). Moreover, the Supreme Court’s decision in Mideal has not removed all the uncertainties in this area, for the parties dispute both the meaning and the application of the standards set forth in Midcal. Finally, sensitive policy considerations influence the choices to be made. On the one hand there is the strong policy in favor of competition that underlies the federal antitrust laws. On the other are the right of the states to engage in economic regulation and all the federalism implications which the potential conflict between state and federal regulation may generate.
In Midcal, after reviewing the line of cases following Parker v. Brown, 317 U.S. 341 , 63 S.Ct. 307 , 87 L.Ed. 315 (1943), the Supreme Court set forth two standards for determining when the state’s involvement is sufficient to justify immunity of the challenged conduct. “First, the challenged restraint must be ‘one clearly articulated and affirmatively expressed as state policy’; second, the policy must be ‘actively supervised’ by the State itself.” 445 U.S. at 105 , 100 S.Ct. at 943 (quoting City of Lafayette v. Louisiana Power & Light Co., 435 U.S. 389, 410 , 98 S.Ct. 1123, 1135 , 55 L.Ed.2d 364 (1978) (opinion of Brennan, J.)). Thus, whether the defendant be a state or state *482 instrumentality or a private party, 2 Midcal directs that the focus for immunity purposes must be upon the extent of the state’s involvement in the challenged restraint— that is, upon the kind of imprint of state authority the anti-competitive activity bears.
Midcal examines the extent of the state’s involvement at two levels — first, at the level of articulation of a state policy in favor of the challenged restraint and second, at the level of implementation of that policy through active supervision. Significant state involvement at both levels is necessary to justify exemption of the anticompetitive conduct under the state action doctrine. Hence the balance that the Supreme Court has struck between the federal policy of competition and the states’ sovereignty is this: the state may choose to displace competition with economic regulation by making clear its intention to do so; once it has made that choice, however, the state bears the burden of ensuring that this policy is furthered and that it remains the ultimate decisionmaker. 3
In testing the actions of private parties under Midcal standards, I am convinced that state compulsion of the private activity, as set forth in Goldfarb, is not a sine qua non for clothing private conduct with state action immunity. It is on this point that the majority and I differ. To state the point more plainly, state compulsion is not the only way to satisfy the first prong of the Midcal test; such compulsion is merely the best evidence that the challenged restraint is a “clearly articulated and affirmatively expressed” state policy.
Support for this interpretation of Midcal may be found in the language of that opinion. The resale price maintenance scheme established by the California statute required of wine wholesalers, on pain of fines or of license suspension or revocation, adherence to the pricing scheme. 445 U.S. at *483 100 , 100 S.Ct. at 940 . In concluding that the first prong of the Midcal test had been satisfied, however, the Court did not track the Goldfarb language and reason that the first prong was met because the state “compelled” resale price maintenance. It chose instead to say: “The legislative policy is forthrightly stated and clear in its purpose to permit resale price maintenance.” Id. at 105 , 100 S.Ct. at 943 (emphasis added).
Moreover, I believe that the function served by the Goldfarb compulsion requirement is served as well by the Midcal formulation. Where private parties are involved to any extent in the state’s scheme of economic regulation, there is properly a concern that the economic self-interest of those parties may infect the decisionmaking process and result in harmful anticompetitive activity. Such private self-interest does not warrant the cloak of state action immunity. By requiring that a state actively supervise the implementation of its economic policy, the Midcal test ensures that the state has determined that the challenged activity is in furtherance of the state’s policy. Hence the second prong of Midcal addresses the concern in Goldfarb by ensuring that the state does not abdicate its responsibility to private parties. 4 When the state ceases to be the real party in interest by abandoning control over a segment of its economy to private parties, the sovereignty of the state cannot be said to be impaired by withholding state action immunity. 5 Thus the Mid-cal test preserves to the states the freedom to pursue state regulatory policies free of antitrust sanctions without at the same time permitting the purely economic self-interest of private parties to disrupt the forces of competition.
Finally, Professor Areeda’s views are in accord with the interpretation I have placed on Midcal’s effect on the Goldfarb compulsion requirement. In discussing the Goldfarb and Cantor cases, Areeda has commented:
Compulsion ... should not be read as an independent requirement for state action immunity; rather, its presence or absence should serve only as strong evidence about state intent. Admittedly, however, some doctrinal confusion exists on this point.
The Supreme Court and lower courts have not applied the compulsion language literally. In Midcal ...., the Court defined the criteria for immunity not in terms of compulsion but in terms of supervision and articulated state policy; the emphasis on supervision implies public scrutiny, deliberation, and review, but not command.
Areeda, Antitrust Immunity for “State Action” After Lafayette, 95 Harv.L.Rev. 435, 438 & n.19 (1981) (citation omitted) (emphasis added). He went on to explain the result in Goldfarb and Cantor (no immunity) as hinging not on absence of compulsion but on absence of meaningful state participation in the challenged conduct. Id. n.19.
*484 In applying the Midcal test as I understand it to the facts of this case, I think it clear that at least the states of North Carolina, Tennessee, and Georgia have policies favoring collective rate formulation by the rate bureaus. 6 As to the remaining states, I would remand the case to the district court for further development of the factual record and for evaluation under the Midcal test.
By legislative enactment the state of North Carolina has clearly endorsed collective rate formulation as a means for achieving the goals of its public service commission. Section 62-152.1(b) of the North Carolina General Statutes provides:
For the purpose of achieving a stable rate structure it shall be the policy of this State to fix uniform rates for the same or similar services by carriers of the same class. In order to realize and effectuate this policy and regulatory goal any carrier subject to regulation by this commission and party to an agreement between or among two or more carriers relating to rates, fares, classifications, divisions, allowances or charges ... or rules and regulations pertaining thereto, or procedures for the joint consideration, initiation or establishment thereof, may under such rules and regulations as the commission may prescribe, apply to the commission for approval of the agreement ....
The effect of such commission approval is set out in section 62-152.1(h).
Parties to any agreement approved by the commission under this section and other parties are ... hereby relieved from the operation of the antitrust laws with respect to the making of such agreement, and with respect to the carrying out of such agreement in conformity with the terms and conditions prescribed by the commission.
These North Carolina provisions parallel the provisions of the Interstate Commerce Act which exempt from federal antitrust laws agreements concerning collective ratemaking among motor carriers operating in interstate commerce. Moreover, they unmistakably indicate that collective rate formulation is part of the regulatory policy of the state of North Carolina. I would therefore hold the joint ratemaking activities of the rate bureaus in that state immune.
Likewise, I would hold the activities of the appellants in Tennessee immune under the state action doctrine. By express commission regulation Tennessee has made clear its intention to permit and encourage collective rate formulation. Rule 1220-2-1-.40 of the Tennessee Public Service Commission states:
The rules and regulations governing the agreements between or among two or more carriers relating to rates, fares, classifications, divisions, allowances, or charges (including charges between carriers and compensation paid or received for the use of facilities or equipment), or rules and regulations pertaining thereto, or procedures for the joint consideration, initiation or establishment thereof as set forth by the Interstate Commerce Commission in § 5a of the Interstate Commerce Act [now codified as 49 U.S.C. § 10706 (b)], are hereby adopted by the Tennessee Public Service Commission; provided, however, that the final determination on any rates, fares, classifications, divisions, allowances, charges, rules and regulations or procedure shall be left in its final determination to the Tennessee Public Service Commission.
In addition, Tennessee has reiterated its policy in favor of collective action by the carriers by a 1980 amendment to its Motor Carrier Act which vests the Tennessee Public Service Commission with power to establish collective ratemaking procedures. Tenn.Code Ann. § 65-1506 (Cum.Supp.1981). All carriers who are party to a collective ratemaking agreement must comply with these procedures.
With respect to Georgia, that state has recently amended its Motor Common Carrier Act of 1931 to require its public service commission to “establish ratemaking procedures for all motor common carriers, which procedure shall include, but not be limited to, collective ratemaking procedures for the joint consideration, initiation and establishment of such rates and charges.” Ga.Code Ann. § 68-613 (Cum.Supp.1981). All motor common carriers subject to the commission’s authority are “required to comply with such ratemaking procedures.” Id. This legislative pronouncement convinces me that state action immunity would be available to the appellants’ activities in Georgia.
The state of Alabama presents a more difficult question. In the proceedings in the district court, Alabama referred to a *485 1942 order of its public service commission as evidence that collective rate formulation is a clearly articulated and affirmatively expressed state policy. On April 20, 1942, the Alabama Public Service Commission ordered that certain motor carriers “prepare, publish and file with the Commission an individual tariff, or ... participate in an agency issue.” Record at 53. It is unclear whether “agency issue” indicates that carriers may publish their tariff through an agent or that carriers may jointly formulate the rates to be proposed in their tariffs. The term is not defined in the record. I would therefore remand to the district court for further development of the record to determine whether the rules and regulations of, or the enabling legislation governing, the Alabama Public Service Commission clearly articulate a policy in favor of collective formulation by carriers.
Finally, as with Alabama, whether the policy of the state of Mississippi conforms to the “clearly articulated and affirmatively expressed” standard in Midcal is unclear from the record developed prior to this appeal. A review of the statutory scheme, Miss.Code Ann. §§ 77-7-1 to -341 (1972), contemplates cooperation among the carriers to the limited extent of establishing joint rates. This falls short, however, of a clear policy in favor of joint formulation of any proposed rates. Nor did Mississippi present any rules or regulations of its public service commission which might reveal such a policy. I would therefore remand for further development of the record as to Mississippi and for decision by the district court in light of what I deem to be the appropriate construction of Midcal.
For the above reasons, I dissent.
. Midcal was decided after the district court had entered final judgment in this case. Hence the district court did not have the benefit, as do we, of the Supreme Court’s pronouncement on this subject.
. In announcing its two-pronged test in Midcal, the Court purported to derive it from the Parker line of cases, citing cases involving private defendants, Goldfarb v. Virginia State Bar, 421 U.S. 773 , 95 S.Ct. 2004 , 44 L.Ed.2d 572 (1975), and Cantor v. Detroit Edison Co., 428 U.S. 579 , 96 S.Ct. 3110 , 49 L.Ed.2d 1141 (1976), as well as cases involving governmental defendants, Bates v. State Bar of Arizona, 433 U.S. 350 , 97 S.Ct. 2691 , 53 L.Ed.2d 810 (1977), and New Motor Vehicle Bd. v. Orrin W. Fox Co., 439 U.S. 96 , 99 S.Ct. 403 , 58 L.Ed.2d 361 (1978).
. There may be substantive limits on what conduct the state may immunize by making the challenged restraint a state policy. Indeed, there is some suggestion in Supreme Court precedent that federal antitrust laws do place substantive limits on the content of state economic regulation. See, e.g., Schwegmann Bros. v. Calvert Distillers Corp., 341 U.S. 384 , 71 S.Ct. 745 , 95 L.Ed. 1035 (1951) (invalidating a Louisiana fair trade statute containing a “non-signer” provision under which a retailer could be enjoined from knowingly underselling another retailer whose prices were set under a resale price maintenance contract, exempt from antitrust laws under the Miller-Tydings Act, with a supplier). But see 1 P. Areeda & D. Turner, Antitrust Law 70 (1978) (interpreting this decision as an inadequate supervision case). See generally Posner, The Proper Relationship Between State Regulation and the Federal Antitrust Laws, 49 N.Y.U.L.Rev. 693 (1974). Those limits may be reached in areas, unlike the private utility regulation involved here, that have not traditionally been regulated by the states.
Fortunately, we are not called upon here to delineate those boundaries. At least one member of the Supreme Court and some commentators have urged an express preemption analysis, which involves the difficult weighing of the state interest at stake against the federal interest in a competitive economy. See Cantor v. Detroit Edison Co., 428 U.S. 579, 605 , 96 S.Ct. 3110, 3124 , 49 L.Ed.2d 1141 (1976) (Blackmun, J., concurring): Posner, supra; Note, Parker v. Brown Revisited: The State Action Doctrine After Goldfarb, Cantor, and Bates, 77 Colum.L.Rev. 898, 929-30 (1977) (suggesting that a preemption analysis explains the Court’s result in some of the cases in the Parker line). But see Cantor, 428 U.S. at 627 , 96 S.Ct. at 3134 (Stewart, J., dissenting) (warning that a preemption approach is a return to the discarded substantive due process doctrine). Cf. Community Communications Co. v. City of Boulder, — U.S. —, —, 102 S.Ct. 835, 845 , 70 L.Ed.2d 810 (1982) (Rehnquist, J., dissenting) (arguing that state action should be analyzed as a preemption problem). The Court, however, has never forthrightly adopted this approach, and, for the present, it appears to have resolved the tension between federalism and antitrust concerns on the side of federalism, albeit with the requirement of active state supervision of the anticompetitive activity. Midcal thus reiterates the federalism concerns at the heart of Parker .
. Professor Areeda has suggested that Mid-cal ’s active supervision requirement is inapplicable to governmental defendants. Areeda, Antitrust Immunity for “State Action" After Lafayette, 95 Harv.L.Rev. 435, 445 n.50 (1981) (“A few courts erroneously appear to use the Midcal formula (clearly articulated state policy plus active supervision of private parties) to require state supervision of governmental defendants.”). In Community Communications Co. v. City of Boulder, the Supreme Court expressly declined to decide whether the active supervision criterion must be met by a municipality with respect to the challenged ordinance. — U.S. —, — n.14, 102 S.Ct. 835 , 841 n.14, 70 L.Ed.2d 810 (1982). If that criterion serves the function identified above, Professor Areeda’s suggestion would appear correct.
. The existence of a state action immunity enables states, like the federal government itself, to define areas inappropriate for market control. Moreover, the adequate supervision criterion ensures that state-federal conflict will be avoided in those areas in which the state has demonstrated its commitment to a program through its exercise of regulatory oversight. At the same time, it guarantees that when the Sherman Act is set aside, private firms are not left to their own devices. Rather, immunity will be granted only when the state has substituted its own supervision for the economic constraints of the competitive market.
1 P. Areeda & D. Turner, Antitrust Law 73 (1978) (footnotes omitted).
. As indicated by the majority, see note 5 supra in the majority opinion, the government did not contend on appeal that the active supervision requirement has not been satisfied.