Appellees Eliot Spitzer and Arthur Roth (collectively "the state") petition the panel for rehearing to reconsider its decision in Freedom Holdings v. Spitzer, 357 F.3d 205 (2d Cir.2004). Familiarity with that decision is assumed, and we will use the acronyms and abbreviations employed in that opinion without explanation here. The state makes various arguments that were, save for one, discussed in the prior opinion. One argument, however, proffers an interpretation of the MSA that was not discussed in our opinion because it was not made. Ordinarily, we would not address an argument made for the first time in a petition for rehearing. Anderson v. Branen, 27 F.3d 29, 30 (2d Cir.1994). However, in light of the apparent importance of this litigation — the petition states that nationwide the states have received $28.6 billion in revenue from the MSA, New York receiving $3.7 billion, see Petition at 3 — and the fact that the anticompetitive effects of the MSA involved disputed issues of fact for trial, we believe that discussion of this new argument is appropriate. Because we find the need to write on this matter, we will take the opportunity to comment on some of the state's other arguments. However, we deny the petition.

The district court dismissed the complaint under Rule 12(b)(6), and we therefore accept as true the material facts alleged in the complaint and draw all reasonable inferences in the plaintiffs' favor. Hernandez v. Coughlin, 18 F.3d 133, 136 (2d Cir.1994). Nevertheless, we may, as we did in our earlier opinion, view those allegations in light of the full terms of the Escrow and Contraband Statutes (collectively "Challenged Statutes") and the MSA. Kramer v. Time Warner Inc., 937 F.2d 767, 774 (2d Cir.1991).

I. The Complaint Adequately Alleges that the MSA and the Challenged Statutes Give the OPMs De Facto Price-Setting and Market-Sharing Authority

The complaint alleges that the Challenged Statutes, in conjunction with the MSA, give the OPMs power to set prices and share markets in the tobacco industry. See 357 F.3d at 215-16. It is alleged, as described below, that NPMs are deterred from seeking increased market share because the high costs of compliance with the Escrow Statute preclude their competing through lower prices. SPMs face heavy monetary penalties for gaining market share and must therefore follow the OPMs' prices to avoid doing so. OPMs must also pay if their market share increases.

However, the petition for rehearing argues that the terms of the MSA show the complaint's allegations to be incorrect and that neither SPMs nor OPMs are deterred from engaging in price competition by substantial cost increases required by the MSA for increases in market share, amounting to penalties. In essence, it argues that the complex market share arrangements of the MSA have no purpose other than to impose a flat levy per cigarette sold and are essentially a superfluous maze leading to a result that could have been achieved in a simple, straight-forward manner. Construing the complaint's allegations most favorably to the plaintiffs, we disagree.

According to the complaint, the function of the Escrow Statute is to coerce NPMs to join the MSA because the costs of compliance with the Escrow Statute are substantially higher than the costs of being an SPM. The plaintiffs allege that the costs of compliance with the Escrow Statute are higher than the costs of joining the MSA because the escrow payments are not tax deductible, unlike MSA payments by PMs. The state has ignored the allegation of taxation effects in the petition for rehearing, reason enough for us to deny the petition.

The state further claims that the costs of compliance with the MSA as an SPM and the costs of compliance with the Escrow Statute as an NPM are the same — 2-cents-per-cigarette. Petition at 14 n. 4. Compliance with the Escrow Statute does cost slightly less than 2-cents-per-cigarette, N.Y. Pub. Health Law § 1399-pp(2)(a), but there is no support in the MSA or in the complaint for the claim that SPMs pay 2-cents-per-cigarette. The MSA provides that an SPM "shall have payment obligations under [the MSA] only in the event that its Market Share in any calendar year exceeds the greater of (1) its 1998 Market Share or (2) 125 percent of its 1997 Market Share." MSA IX(i)(1). Thus, with one qualification, we must at this stage view the costs of complying with the Escrow Statute as higher than the costs of becoming an SPM and maintaining market share.

The qualification concerns the MSA's definition of 1997 and 1998 market share. Any manufacturer that becomes an SPM more than sixty days after the MSA execution date is considered to have had zero market share in 1997 and 1998. MSA IX(i)(4). Thus, such SPMs must make all of the annual payments made by OPMs according to their market share — though they are not required to make any of the initial payments required of the OPMs. MSA IX(i)(1). There is no indication in anything before us that these payments amount to 2-cents-per-cigarette.

The alleged result regarding the scheme's effect on manufacturers and their decisions to join or not join the MSA is that: (i) NPMs pay more in escrow payments than SPMs, partially due to tax considerations; (ii) SPMs that join more than 60 days after the MSA execution date make annual payments based on their market shares, just like OPMs; and (iii) SPMs that join within 60 days of the execution date pay nothing, unless their market shares increase above 1998 (or 125% of 1997) levels. Obviously, if these allegations are true, a manufacturer would prefer to be in group (iii) over group (ii) and group (ii) over group (i), and the scheme is allegedly coercive for this reason. It is also notable that the time limit for joining group (iii) was only sixty days — arguably not enough time to allow a cigarette manufacturer to evaluate its options in light of actual experience under the MSA as enforced by the states.

As alleged, therefore, the structure of the MSA and enforcing statutes causes the manufacturers that become SPMs to follow the price increases of the OPMs to avoid gaining market share. This is especially true for the SPMs that joined within 60 days of the execution date; they have no annual payment obligations and can avoid incurring them by not gaining market share. The SPMs that joined after 60 days must also avoid gaining market share. If they take market share from the OPMs, they will be forced to pay proportionately more of the annual payments to the states. This is because the percentage SPMs pay is determined by the following quotient:

current market share —
1998 (or 125% of 1997) market share
current aggregate market share of
OPMs.

MSA IX(i)(2). Because 1997 and 1998 market share are defined as zero for these SPMs, see supra, the quotient is effectively "current market share" divided by "current aggregate market share of OPMs." If the denominator were current aggregate market share of OPMs and SPMs, gaining market share from OPMs would be less harmful for SPMs, because the denominator would not change even when the numerator increased. But under the MSA, if the numerator increases because the SPM has taken market share from an OPM, the denominator decreases by the amount of the increase. Thus, the SPM's proportion of the annual payment increases by more than its proportion of overall market share.

The petition states that an SPM's "marginal payment per cigarette is always lower than an OPM's per-cigarette payment," Petition at 10 n. 1. This statement is somewhat ambiguous but, again viewing the allegations in the light most favorable to appellants, appears to be incorrect. Where the SPM's 1997 and 1998 market share are defined as zero under the agreement, the SPM makes the same payments per cigarette as the OPM in year one, and if the SPM takes market share from the OPM in year two, the SPM pays more per cigarette than the OPM that year. MSA IX(i)(1)-(2). Only SPMs that join the MSA within 60 days of its execution date and do not gain market share are certain to pay less per-cigarette — nothing — than OPMs. MSA IX(i)(1), (4). Moreover, the central issue is not absolute payment numbers but relative numbers and whether payments increase disproportionately (i.e. in more than a 1 to 1 ratio) when market share increases.

As alleged, the end result is as follows: (i) NPMs will lose by charging lower prices because the costs of complying with the Escrow Statute are too high; (ii) SPMs that joined within 60 days of the execution date similarly lose if they charge lower prices than OPMs because if their market shares increase above 1998 levels or 125% of 1997 levels, they will have to make annual payments and will then be unable to compete on price for the reasons listed next; (iii) SPMs that joined after 60 days of the execution date (or that joined earlier but gained market share) cannot compete on price because if they steal market share from the OPMs, their annual payments will increase disproportionately, and the increased market share will (allegedly) not be profitable.

Given the allegations of the complaint, we cannot dismiss a challenge to this scheme on the grounds that it is not, as a matter of law, anti-competitive. For reasons stated in our earlier opinion and here, the complaint alleges such an effect and plaintiffs are entitled to attempt to prove it.

The petition argues that Midcal1 is inapplicable because: (i) the challenged statutes constitute a unilateral act of state and are therefore insulated from invalidation under the antitrust laws, despite any anticompetitive effects; (ii) the panel has misunderstood the MSA, which is not a marketsharing agreement among manufacturers; and (iii) the statutes do not delegate any regulatory authority to private parties.

These arguments are meritless under either of two theories. First, although the statutes themselves are acts of the state, their function is to enforce the MSA, a market-sharing agreement between private parties. 357 F.3d at 224-25; Section I, supra. Thus, the MSA is part and parcel of the Challenged Statutes, which are accordingly required to meet the Midcal test to be protected by Parker2 immunity.

Second, unilateral acts of state such as the statutes at issue can be challenged under Section 1 of the Sherman Act as hybrid restraints where they grant to "private actors" "a degree of private regulatory power." 324 Liquor Corp. v. Duffy, 479 U.S. 335, 345 n. 8, 107 S. Ct. 720, 93 L. Ed. 2d 667 (1987) (internal quotation marks and citations omitted); 357 F.3d at 223.

The petition argues that "the only function of the Challenged Statutes is to require cigarette manufacturers to certify compliance with the [Escrow Statute's] requirement that they internalize health costs imposed by their cigarettes, either by joining the MSA or by maintaining the required escrow. Private parties must abide by the Challenged Statutes ... but no authority is delegated to them." Petition at 12. This argument is meritless.

First, the argument by its terms concedes that the statutes delegate regulatory authority to private parties. As the state points out, the statutes require a manufacturer to either join the MSA or to comply with the Escrow Statute. A manufacturer who does either is forced, according to the allegations of the complaint, to become part of the market-sharing agreement set up by the MSA — i.e. it must not gain market share and it therefore cannot compete on price (or, in the case of NPMs, they allegedly cannot afford to do either). See Section I, supra. Because the petition does not assert that the OPMs' prices are subject to supervision by the state, see Section IV infra, the Challenged Statutes have the effect, according to the allegations of the complaint, of delegating price-setting authority to the OPMs.

III. The State Misinterprets Our Discussion of the Ancillary Purpose of the First Midcal Prong

The first Midcal prong states that "the challenged restraint must be one clearly articulated and affirmatively expressed as state policy." 445 U.S. at 105, 100 S. Ct. 937 (internal quotations omitted). The main requirement of this prong is that the restraint be an express act of the state, a requirement met in this case. 357 F.3d at 227. The opinion also states that this prong contains an ancillary purpose — "to reveal the State's purposes in agreeing to, and enforcing," the restraint. Id. at 227-28.

This ancillary purpose is included in the Midcal analysis because of Parker: because the state's "purposes must be known to ensure that the State's policy goals are sufficient to qualify for the Parker immunity — simply protecting private parties from competition is not a sufficient goal." 357 F.3d at 227. As Midcal itself stated, a court must ensure that "[t]he national policy in favor of competition cannot be thwarted by casting ... a gauzy cloak of state involvement over what is essentially a private price-fixing arrangement." Midcal, 445 U.S. at 106, 100 S. Ct. 937; see also Parker, 317 U.S. at 351, 63 S. Ct. 307 ("a state does not give immunity to those who violate the Sherman Act by authorizing them to violate it, or by declaring that their action is lawful") (internal citations omitted).

Indeed, the opinion notes that the Parker analysis and the ancillary purpose of the Midcal prong are interchangeable. 357 F.3d 227 ("if the purposes are not of the kind that would trigger Parker analysis, we generally would deny the immunity on Parker grounds rather than on a failure to satisfy the first Midcal prong"). Furthermore, the opinion conceded that failure to meet the ancillary purpose test alone — would not be enough to upset a statue. Id.

The petition of course does not argue that Parker is irrelevant. Rather, it claims that there is no ancillary purpose because under Town of Hallie v. City of Eau Claire, the first Midcal prong is satisfied if suppression of competition is the "`foreseeable result'" of what the statute authorizes. Petition at 13 n. 3 (quoting Hallie, 471 U.S. 34, 42, 105 S. Ct. 1713, 85 L. Ed. 2d 24 (1985)). Hallie, however, discusses the ability of a municipality to engage in anticompetitive schemes with state authority and expressly notes that such cases are distinct from cases involving private parties "because a municipality is an arm of the State" and "[w]e may presume, absent a showing to the contrary, that the municipality acts in the public interest." 471 U.S. at 45, 105 S. Ct. 1713.

The petition also relies upon FTC v. Ticor Title Insurance Co. for the proposition that "`[i]n the usual case, Midcal's requirement that the State articulate a clear policy shows little more than that the State has not acted through inadvertence.'" Petition at 13 n. 3 (quoting Ticor, 504 U.S. 621, 636-37, 112 S. Ct. 2169, 119 L. Ed. 2d 410 (1992)). This phrase is taken out of context, and supports the panel's opinion in any event. The full quotation is as follows:

For States which do choose to displace the free market with regulation, our insistence on real compliance with both parts of the Midcal test will serve to make clear that the State is responsible for the price fixing it has sanctioned and undertaken to control.

Respondents contend that these concerns are better addressed by the requirement that the States articulate a clear policy to displace the antitrust laws with their own forms of economic regulation. This contention misapprehends the close relation between Midcal's two elements. Both are directed at ensuring that particular anticompetitive mechanisms operate because of a deliberate and intended state policy. In the usual case, Midcal's requirement that the State articulate a clear policy shows little more than that the State has not acted through inadvertence; it cannot alone ensure, as required by our precedents, that particular anticompetitive conduct has been approved by the State.

Ticor, 504 U.S. at 636-37, 112 S. Ct. 2169 (internal citation omitted). In this statement, the Court was referring to judicial oversight of individual private parties and held that requiring clear articulation alone would not suffice to ensure that private acts were approved by the state, because the state could not know in advance how its articulation would be interpreted. Thus, to satisfy itself that the particular anticompetitive act of an individual furthered state policy, a court would also have to know that the state was aware of that party's acts and was supervising them.

The fact that clear articulation alone is insufficient to confer state-action immunity, however, says little about the purposes of requiring clear articulation or the uses to which such an articulation may be put. To be sure, the court must find under this prong that the state did not inadvertently include anticompetitive activities in some larger scheme. Id. For this reason, it is important that a state enunciate its intent to displace competition when it means to do so. However, it is also obvious that a state's explanation of its reason for displacing competition will aid a court in determining whether the state has satisfied the requirements for Parker immunity. Aiding judicial inquiry in this way therefore represents an ancillary purpose of the first Midcal prong.

Parker bans states from exempting private parties from the restrictions of the Sherman Act solely to benefit those parties. 317 U.S. at 351-52, 63 S. Ct. 307. That ban would be wholly ineffectual if such a scheme could be legitimated solely by reference to some state goal with no plausible nexus to the exemption. For example, a court facing an antitrust challenge to a state statute allowing price-fixing by car washes is not obliged to dismiss the complaint because the state asserts that the law will improve the performance of the state capital city's symphony orchestra.

The petition itself nicely illustrates the issue. It states that the Challenged Statutes further public health, force cigarette manufacturers to "internalize health costs imposed by their cigarettes," and "ensure that those who cause the health crisis pay for it." Petition at 3, 4, 12. But it never states how the alleged market-sharing scheme furthers those goals. It simply denies that the market-sharing scheme is anticompetitive, leaving the purpose of this complex, carefully drafted set of rules an enigma.

In this regard, the petition points to the portion of the opinion stating that Parker immunity is shown only where the state acted "in furtherance of legitimate state policy goals and limits unnecessary anticompetitive effects." 357 F.3d at 223; Petition at 13-14. We agree that the statement regarding "unnecessary anticompetitive effects" goes further than is necessary to our present decision, whether or not it flows from Parker and Midcal. We need say only that a state must act in furtherance of legitimate state policy goals and its acts must have a plausible nexus to those goals.

The remainder of the petition's arguments concerning the ancillary purpose section are meritless. The petition correctly argues that courts cannot use the first Midcal prong to second-guess whether the state goal is in the "public interest," Petition at 14, but the opinion does not purport to go so far and states only that under the first Midcal prong, the state goal must be "legitimate." 357 F.3d at 230. The petition of course does not argue that state goals need not be legitimate.

IV. The Opinion Expressly Relies on the Second Rather than the First Midcal Prong

As suggested by the fact that we found it "doubtful that a federal court would upset a state statute solely because it failed to meet the explanatory aspect of the first Midcal prong," 357 F.3d at 227, the opinion expressly relied on the challenged scheme's failure to meet the second Midcal prong as its justification for finding that the statutes are not shielded from the antitrust laws by Parker immunity so as to permit dismissal on the allegations of the complaint. 357 F.3d at 231. Notably, the petition does not claim that the active supervision requirement was met. It is too soon to say whether the state will ultimately be able to elicit evidence sufficient to meet this second prong.

V. The MSA Signatories Sought an Antitrust Exemption for an Earlier Agreement Among those Parties, but because the Market-Sharing Provisions of the Earlier Agreement are Similar to or Less Anticompetitive than those of the MSA, Discussion of the Antitrust Exemption Hearings Remains Relevant

The petitioners correctly note that an antitrust exemption was sought from Congress for a different agreement among the tobacco manufacturers and the states — not the MSA but rather an agreement to seek a legislative settlement of suits by the states against the tobacco companies. This "Global Settlement Agreement" ("GSA") was transmuted into the "National Tobacco Policy and Youth Smoking Reduction Act," S. 1415, 105th Cong. (1987).3 If the opinion was faulty in not noting this distinction, the omission is immaterial because the two agreements included similar market-sharing schemes. In fact, the market-sharing provisions in the GSA may have been less anticompetitive than those alleged to be in the MSA, and thus the Senate's discussions of the anticompetitive effects of the former are fully applicable to the latter.

The bill required annual payments to a fund in perpetuity by all participating manufacturers. Id. at § 403(b). These payments were apportioned among participating manufacturers according to their market share in the most recent calendar quarter. Id. at § 403(d). Thus, under the bill (like the MSA), a manufacturer's payment obligations varied with its market share.

The bill included provisions aimed at preventing small manufacturers from competing on price and gaining market share. A small manufacturer would receive financial benefits if it kept its sales below certain thresholds; if its sales rose above 500 million "units" per year, it lost all benefits. Id. at § 403(d)(3).

A "pass-through" provision of the bill provided that "each [PM] shall use its best efforts to adjust the price at which it sells each unit of tobacco products ... by an

amount sufficient to pass through to each purchaser on a per-unit basis an equal share of the annual payments to be made by such [PM]." Id. at § 405(a). Failure to abide by this provision would subject a PM to a fine equal to between 115% and 125% of the undercharge. Id. at § 405(b)(1). Thus, any competition based on price would subject a manufacturer to charges that it was not passing through the annual payments, the resolution of which would be costly even if the undercharge was determined not to have occurred.

Finally, NPMs would have been subject to heavy payment obligations under the proposed bill. Each NPM had to pay 150% of what it would have paid annually under Section 403 of the bill as a "fee." Id. at § 708(b). The corresponding payments made by PMs were tax deductible, but these NPM payments were not. See id. at § 406. Thus, there would have been significant pressure on NPMs to become PMs, and, as noted, the PMs were constrained in their ability to compete by the provisions described above.

We therefore deny the petition for rehearing.