LYNCH, Chief Judge.

The plaintiffs in this case complain that the prices for gasoline on Martha's Vineyard have been artificially high due both to an illegal price-fixing conspiracy among four of the island's nine gas stations and to unconscionable price-gouging in the aftermath of Hurricanes Katrina and Rita in 2005. As to the antitrust claims, the stations agree for the purposes of summary judgment that there is evidence of parallel pricing but say that is not illegal absent an agreement to fix prices. The district court granted summary judgment to defendants on both of plaintiffs' claims, which were brought under § 1 of the Sherman Anti-Trust Act and a price-gouging regulation under Mass. Gen. Laws ch. 93A. We affirm, discussing the law on "agreements," as opposed to "conscious parallelism," under the Sherman Act, and assessing the defendants' post-hurricane pricing patterns under the state price-gouging rule.

I. Standard of Review

We discuss separately the price-fixing and price-gouging claims. The standard of review for each is the same. We review the district court's grant of summary judgment de novo, taking all facts and reasonable inferences therefrom in the light most favorable to plaintiffs, the nonmoving parties, and affirming only if there are no genuine issues of material fact and defendants are entitled to judgment as a matter of law. See Cortes-Rivera v. Dep't of Corr. and Rehab., 626 F.3d 21, 26 (1st Cir.2010).

II. Sherman Act Price-Fixing Claim

An understanding of the legal structure of a price-fixing claim under the Sherman Anti-Trust Act gives context to the facts relied on by plaintiffs on summary judgment.

A. Legal and Economic Background

Section 1 of the Sherman Anti-Trust Act prohibits "[e]very contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade." 15 U.S.C. § 1.1 In general, practices challenged under the Sherman Act are struck down only if they are unreasonable and anticompetitive, but agreements to fix prices are "so plainly anticompetitive" that they are per se illegal. Texaco Inc. v. Dagher, 547 U.S. 1, 5, 126 S.Ct. 1276, 164 L.Ed.2d 1 (2006) (quoting Nat'l Soc'y of Prof'l Eng'rs v. United States, 435 U.S. 679, 692, 98 S.Ct. 1355, 55 L.Ed.2d 637 (1978)).

Section 1 by its plain terms reaches only "agreements"—whether tacit or express. Bell Atl. Corp. v. Twombly, 550 U.S. 544, 553, 127 S.Ct. 1955, 167 L.Ed.2d 929 (2007). It does not reach independent decisions, even if they lead to the same anticompetitive result as an actual agreement among market actors.2 15 U.S.C. § 1; Am. Needle, Inc. v. Nat'l Football League, ___ U.S. ___, 130 S.Ct. 2201, 2208-09 & n. 2, 176 L.Ed.2d 947 (2010); Clamp-All Corp. v. Cast Iron Soil Pipe Inst., 851 F.2d 478, 484 (1st Cir.1988). The statute "does not require sellers to compete; it just forbids their agreeing or conspiring not to compete." In re Text Messaging Antitrust Litig., 630 F.3d 622, 627 (7th Cir.2010) (Posner, J.).

This limit means that bare "conscious parallelism" is "not in itself unlawful." Brooke Grp. Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 227, 113 S.Ct. 2578, 125 L.Ed.2d 168 (1993). Conscious parallelism3 is a phenomenon of oligopolistic markets4 in which firms "might in effect share monopoly power, setting their prices at a profit-maximizing, supracompetitive level by recognizing their shared economic interests and their interdependence with respect to price and output decisions." Id. Each producer may independently decide that it can maximize its profits by matching one or more other producers' price, on the hope that the market will be able to maintain high prices if the producers do not undercut one another.

A tacit agreement—one in which only the conspirators' actions, and not any express communications, indicate the existence of an agreement—is distinguished from mere conscious parallelism by "uniform behavior among competitors, preceded by conversations implying that later uniformity might prove desirable or accompanied by other conduct that in context suggests that each competitor failed to make an independent decision." Brown v. Pro Football, Inc., 518 U.S. 231, 241, 116 S.Ct. 2116, 135 L.Ed.2d 521 (1996) (internal citations omitted). In the seminal case, Interstate Circuit v. United States, 306 U.S. 208, 59 S.Ct. 467, 83 L.Ed. 610 (1939), the Supreme Court found a tacit agreement where a dominant movie theater company sent a letter openly addressed to all eight major national film distributors stating that it would show a distributor's films only if the distributor imposed certain restrictions on later runs of the films in secondary theaters. Id. at 215-19, 59 S.Ct. 467. The Supreme Court held that the distributors, who never communicated directly with one another, nonetheless had entered into a tacit agreement with one another by acting in accordance with the letter's demands, because the letter made it clear that all eight had received the letter, the economic context made it clear that all eight needed to act uniformly or all would lose business, and all eight did in fact impose the conditions. Id. at 222, 59 S.Ct. 467. The opinion has been criticized, see, e.g., 3B Areeda & Hovenkamp, Antitrust Law ¶ 810b, at 470-71 (3d ed. 2008), but the Supreme Court has recently reiterated that tacit agreements are still agreements under the Sherman Act, Twombly, 550 U.S. at 553, 127 S.Ct. 1955.

Some markets are particularly conducive to maintaining consciously parallel pricing without the need for agreement among the producers. "Tacit coordination is facilitated by a stable market environment, fungible products, and a small number of variables upon which the firms seeking to coordinate their pricing may focus." Brooke Grp., 509 U.S. at 238, 113 S.Ct. 2578. Such coordination is also easier to maintain when these fungible goods "are repeatedly sold in market transactions that are immediately known in every detail by customers and rivals." 6 Areeda & Hovenkamp, Antitrust Law ¶ 1430b, at 225 (3d ed. 2010). A geographically constrained gasoline market with publicly posted prices has these characteristics.

Because supracompetitive prices— prices above what they would be in a perfectly competitive market—can result from both lawful conscious parallelism and an unlawful agreement to fix prices, antitrust doctrine has developed evidentiary standards to minimize the risk that legal conduct will be chilled or punished. Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752, 763, 104 S.Ct. 1464, 79 L.Ed.2d 775 (1984). Plaintiffs must establish that it is plausible that defendants are engaged in more than mere conscious parallelism, by pleading and later producing evidence pointing toward conspiracy, sometimes referred to as "plus factors." See Twombly, 550 U.S. at 556 & n. 4, 127 S.Ct. 1955 (requiring antitrust plaintiffs to plead behavior more consistent with agreement than with independence); In re Flat Glass Antitrust Litig., 385 F.3d 350, 360 (3d Cir.2004) (explaining that "plus factors" are "proxies for direct evidence of an agreement").

In addition, the Supreme Court has "limit[ed] the range of permissible inferences from ambiguous evidence in a § 1 case," holding that, at summary judgment, "conduct as consistent with permissible competition as with illegal conspiracy does not, standing alone, support an inference of antitrust conspiracy" that allows plaintiffs' evidence to reach a jury. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 588, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986).

In order to survive summary judgment, plaintiffs must produce direct or circumstantial evidence that is not only consistent with conspiracy, but "tends to exclude the possibility of independent action." Monsanto, 465 U.S. at 764, 104 S.Ct. 1464. Such evidence could show "parallel behavior that would probably not result from chance, coincidence, independent responses to common stimuli, or mere interdependence unaided by an advance understanding among the parties." Twombly, 550 U.S. at 557 n. 4, 127 S.Ct. 1955 (quoting 6 Areeda & Hovenkamp, Antitrust Law ¶ 1425, at 167 (2d ed. 2003)) (internal quotation marks omitted).

These special rules apply to claims of horizontal conspiracies such as this claim of price-fixing.5 See Twombly, 550 U.S. at 554, 127 S.Ct. 1955 (stating that a § 1 plaintiff must meet the Monsanto and Matsushita requirements, and not distinguishing among types of § 1 claims); see also, e.g., Flat Glass, 385 F.3d at 357-59 (applying Matsushita in price-fixing case, because lawful conscious parallelism can lead to same economic result as conspiracy); Blomkest Fertilizer, Inc. v. Potash Corp. of Saskatchewan, 203 F.3d 1028, 1032 (8th Cir.2000) (stating that circuit applies Matsushita "broadly, and in both horizontal and vertical price fixing cases"). This is the common understanding and plaintiffs do not disagree.

B. Factual and Procedural Background

The underlying facts of the case, including the district court's description of the retail gasoline market on Martha's Vineyard, are undisputed.

Plaintiffs are summer and year-round residents of Martha's Vineyard and an island real estate agency. Defendants operate four of the nine gas stations on Martha's Vineyard. In Edgartown, individual defendant Francis Paciello owns the Edgartown Mobil station. The defendant corporation Depot Corner, Inc., which is wholly owned by Paciello, owns the Depot Corner station, also in Edgartown. In Vineyard Haven, about seven miles northwest of Edgartown, defendant corporation Drake/Kenyon (Drake)6 owns the Xtra-Mart Citgo station in Vineyard Haven and also supplies gasoline to the Edgartown stations at wholesale. Defendant corporation R.M. Packer, also in Vineyard Haven, owns the Tisbury Shell gas station; R.M. Packer is owned by the individuals Ralph Packer and his wife.

Of the other five gas stations on the island, two are in Oak Bluffs, about three miles from Vineyard Haven, one is in Edgartown, and two are on the western part of the island in West Tisbury and Chilmark.

The defendants' prices exceeded prices at gas stations on Cape Cod by an average of fifty-six cents per gallon during a five year period beginning on August 1, 2003, according to calculations performed by plaintiffs' expert and cited by the district court. Twenty-one cents of that difference is attributable to the higher costs of transporting gas to the island than to the mainland Cape.

The retail gasoline market on Martha's Vineyard has features that make it susceptible to efforts by gas stations to sustain supracompetitive prices. These features would facilitate both conspiratorial pricing and merely interdependent parallel pricing for several reasons. See 2B Areeda, Hovenkamp, & Solow, supra, ¶ 404, at 9-20.

First, would-be competitors attracted to the market by high profit margins face a regulatory barrier to entry: they need permission from the Martha's Vineyard Commission. The Commission has denied all petitions to open new gas stations since 1997. This, along with their location on a relatively small island, insulates the current stations from competition.

Second, customer demand for gasoline on an island is inelastic, meaning customers will not buy much less gas when prices rise, because they cannot choose to drive farther away to get cheaper gas. Gasoline in general is a nondurable good, so that customers have to buy it frequently and are not likely to simply stay out of the market until prices drop. This is particularly true for customers who are summer residents and are in the market for only limited periods of time.

Third, gasoline is a homogeneous good, so consumers decide where to buy it based mostly on price and convenience, leading competing gas stations to prominently post prices. This posting also lets competitors know and respond in real time to one another's prices, allowing them to catch price "cheaters" and to follow price "leaders." See id. ¶ 404b3, at 14-15 (describing cheating); 6 Areeda & Hovenkamp, supra, ¶ 1430d, at 226 (describing leading); see also United States v. Heffernan, 43 F.3d 1144, 1149 (7th Cir.1994) (describing catching price cheaters). Since there are only nine gas stations on the entire island, each station can easily monitor and respond to the prices of the others. If one station drops its price in order to attract more business, the others can quickly drop their prices in response. The original "cheater" benefits very little from undercutting its competitors' prices, because when any one of them drops its prices the competitors can match the price before many customers respond to the incentive. And all of the stations suffer a decrease in profit margin.

Conversely, a station acting as a price "leader" risks little by raising its price under such market conditions. Other stations are likely to follow, given the possibility of higher prices and profit margins for all. If for some reason the competitors do not follow the increases, the leader can easily drop its price again to match the other stations so quickly that few customers are lost to lower-priced competition. Knowing these features of the market, each gas station owner is likely to reach its own independent conclusion that its best interests involve keeping prices high, including following price changes by a price "leader" (if one emerges), in confidence that the other station owners will reach the same independent conclusion. Here there is no evidence or suggestion that the business risk to any station on Martha's Vineyard of raising its prices was so great as to require communication among stations before any one of them would venture it.

Plaintiffs, angered by what they saw as unjustifiably high prices at the pump, brought suit in Massachusetts Superior Court on August 2, 2007 alleging that defendants had conspired to fix retail gas prices since at least December 31, 1999. Defendants removed the case to federal court on August 28, 2007,7 after which plaintiffs filed an amended complaint, substituting Sherman Anti-Trust Act price-fixing allegations or their original Massachusetts Antitrust Act allegations. Due to the Sherman Act's four-year statute of limitations, 15 U.S.C. § 15b, the district court limited plaintiffs' price-fixing claim to violations occurring on or after August 2, 2003. The district court allowed discovery, but denied the parties' joint requests for discovery of detailed sales and financial data from other Martha's Vineyard gas stations not named in the suit. Following a June 24, 2009, hearing on defendants' motions for summary judgment, the district court granted defendants' motions on January 6, 2010.

Although our review is de novo, we describe the district court's thoughtful analysis. The district court reasoned that most of the evidence the plaintiffs proffered was no more consistent with conspiracy than with independent choices to engage in parallel pricing. The district court found other evidence not sufficiently probative of conspiracy. First, the district court found that Drake's employment of a consultant to lobby the Martha's Vineyard Commission to deny petitions for new gas stations was "conduct that would be expected even in a competitive gasoline retail market." White v. R.M. Packer Co., No. 07-11601, slip op. at 6 n. 5 (D.Mass. Jan. 6, 2010). Second, the district court found too attenuated the evidence that two of the defendants communicated with one another about prices in 1999 because the communications concerned the wholesale, rather than retail, market and occurred years before the period covered by the limitations period. The district court found it troubling that an unusually generous loan from Drake to Paciello may have provided Paciello with an incentive to conspire with Drake, but held that this was not enough to permit a reasonable jury to find an agreement to fix prices.

C. Merits of Plaintiffs' Appeal from Entry of Summary Judgment

Plaintiffs argue that the district court erred by (1) applying the wrong legal standard, (2) considering their evidence of plus factors piecemeal rather than as a whole, and (3) ignoring their expert evidence.

First, plaintiffs argue that the district court, despite correctly stating the legal standard to which it held plaintiffs' evidence, proceeded in fact as if plaintiffs are required to "exclude," rather than "tend to exclude," the possibility of independent action. See Matsushita, 475 U.S. at 588, 106 S.Ct. 1348. Having reviewed the entire record, we disagree. Nothing in the district court's opinion suggests that the court required plaintiffs to produce direct evidence of conspiracy as plaintiffs claim.8

Plaintiffs also argue that defendants' concession for summary judgment purposes that there has been parallel pricing "gets plaintiffs close to defeating summary judgment" because of the statement in Twombly that "[a]n allegation of parallel conduct ... gets [a] complaint close to stating a claim" for purposes of surviving an initial motion to dismiss. Twombly, 550 U.S. at 557, 127 S.Ct. 1955 (emphasis added). As a statement of the law at summary judgment, plaintiffs are flatly wrong. Mere parallelism, whether stipulated or proven, does not even create a prima facie conspiracy case. See Brooke Grp., 509 U.S. at 227, 113 S.Ct. 2578.

Much of the evidence plaintiffs offer as "plus factors," even when viewed in the light most favorable to them, does no more than corroborate that the Martha's Vineyard gasoline market is an oligopolistic market which is highly conducive to parallel pricing. The evidence does nothing to explain whether the parallel pricing was achieved by agreement or mere interdependent decisions. Plaintiffs' remaining evidence does not "`tend[] to exclude the possibility' that the alleged conspirators acted independently," and so is not enough to permit a reasonable inference that defendants' behavior was more than mere conscious parallelism. Matsushita, 475 U.S. at 588, 106 S.Ct. 1348 (quoting Monsanto, 465 U.S. at 764, 104 S.Ct. 1464). The report of plaintiffs' expert witness, Boston College economics professor Frank Gollop, does not alter the conclusion. The report was adequately considered by the district court. It does not undermine, and in fact is consistent with, our conclusion and that of the district court.

Plaintiffs draw their list of plus factors from the authorities mentioned in the Supreme Court's Twombly opinion as providing "examples of parallel conduct allegations that would state a § 1 claim" under the heightened pleading standard established in that case. Twombly, 550 U.S. at 557 n. 4, 127 S.Ct. 1955 (citing 6 Areeda & Hovenkamp, supra, ¶ 1425, at 167-85; Michael D. Blechman, Conscious Parallelism, Signalling and Facilitation Devices, 24 N.Y.L. Sch. L. Rev. 881, 899 (1979)). Plaintiffs also invoke a list of plus factors described in Judge Posner's antitrust monograph. See Richard Posner, Antitrust Law 69-93 (2d ed. 2001). Yet as these sources themselves emphasize, many so-called plus factors simply "demonstrate that a given market is chronically non-competitive," without helping to explain whether agreement or conscious parallelism is the cause. Blechman, supra, at 898; see also 6 Areeda & Hovenkamp, supra, ¶ 1434, at 263-69 (explaining that "plus factors" which merely restate interdependence are prerequisites to an inference of agreement but are not enough); Posner, supra, at 69-79 (listing seventeen factors as useful in identifying markets conducive to conscious parallelism). To be sure, providing this evidence demonstrates that plaintiffs' claim is not economically implausible. See Matsushita, 475 U.S. at 587, 106 S.Ct. 1348 ("[I]f the claim is one that simply makes no economic sense[, plaintiffs] must come forward with more persuasive evidence to support their claim than would otherwise be necessary."). But such evidence does not by itself suggest that defendants' conduct shows agreement. See Flat Glass, 385 F.3d at 360-61; see also In re High Fructose Corn Syrup Antitrust Litig., 295 F.3d 651, 658 (7th Cir. 2002) (Posner, J.) (distinguishing "evidence of noncompetitive behavior" from "evidence that the structure of the market was conducive to such behavior").

We have considered the totality of plaintiffs' evidence, as did the district court, and discuss the inadequacy of the parts as well as of the whole. Six of plaintiffs' nine plus factors show nothing more than that the gasoline market on Martha's Vineyard is conducive to conscious parallelism. We begin with two that describe defendants' pricing behavior. First, during three periods in 2004 and 2005 the defendants, in parallel, held prices steady or raised them while the cost of gasoline at wholesale declined. Second, defendants have enjoyed what plaintiffs call "abnormal profits." It is true, as plaintiffs state, that changes in pricing patterns and profit levels may be useful in identifying the beginning of a conspiracy, immediately after which conspirators may successfully raise prices without reference to costs. See Posner, supra, at 88, 90; see also Text Messaging, 630 F.3d at 628 (stating that in a competitive market "falling costs ... motivat[e a seller], in the absence of agreement, to reduce his price slightly in order to take business from his competitors").

However, these pricing behaviors do not function as "plus factors" when they are stable over time, because that factual context undermines any inference that the pricing behavior represents a sudden shift marking the beginning of a price-fixing conspiracy. Here, plaintiffs allege a price-fixing conspiracy dating back "at least as early as December 31, 1999"; there is no suggestion that defendants' behavior changed in 2004, in 2005, or at any other relevant time. In this factual context, evidence that defendants set supracompetitive prices that did not decline when their costs declined shows no more than that they made their pricing decisions in an oligopolistic, rather than competitive, market.

Other evidence also undermines plaintiffs' arguments about defendants' pricing behaviors. Plaintiffs' own expert stated that "[n]o unambiguous conclusion ... can be gleaned from [defendants'] pattern of parallel pricing," and that he could "draw no behavioral conclusion" from the divergence between the defendants' declining costs and retail prices, since it could "be the result of a supra-competitive price umbrella." Further, the expert did not explain whether in his opinion such an "umbrella" would even require agreement. Similarly ambiguous is deposition testimony as to three of the four stations, by the principals of defendants Drake and Depot Corner, Inc., that they did not know what margin over cost they needed to charge to turn a profit. Plaintiffs argue that without a price agreement such ignorance would be "suicidal," but the testimony equally supports the inference that stations' pricing decisions in this market are not based, as they would be in a competitive market, on cost, but rather on the actions and expected actions of other stations.

A third plus factor offered by plaintiffs is that defendants had "motive to conspire" because cooperating with one another could allow them to earn supracompetitive profits. Taking as a given that all of the defendants had motive to conspire with one another to earn high profits, all such a motive shows is that the defendants could reasonably expect to earn higher profits by keeping prices at a supracompetitive level through parallel pricing practices. Matsushita states that evidence showing defendants have "a plausible reason to conspire" does not create a triable issue as to whether there was a conspiracy. Matsushita, 475 U.S. at 596-97 & n. 21, 106 S.Ct. 1348; 6 Areeda & Hovenkamp, supra, ¶ 1434(c)(1), at 269 ("Motivation is thus synonymous with interdependence and therefore adds nothing to it."). We take up later the issue of whether a $2 million loan at a below-market interest rate from Drake to Paciello, when Paciello could not obtain a loan from other lenders, gave Paciello motive to "do Drake's bidding regarding keeping up gas prices" as plaintiffs claim.

Three other plus factors plaintiffs allege are that the Martha's Vineyard retail gasoline market (1) is insulated by high barriers to entry, thanks to the Martha's Vineyard Commission's demonstrated gatekeeping power, (2) faces highly inelastic demand for gasoline, and (3) is marked by stable relative market shares over time among the four defendants. High barriers to entry and inelastic demand are two hallmarks of oligopolistic markets susceptible to successful parallel pricing practices, but neither helps to distinguish between agreement and mere conscious parallelism as the root cause of those practices. "[W]ithout barriers to entry it would presumably be impossible to maintain supracompetitive prices. ..." Matsushita, 475 U.S. at 591 n. 15, 106 S.Ct. 1348; see also Williamson Oil Co. v. Philip Morris USA, 346 F.3d 1287, 1317 (11th Cir.2003) (describing high barriers to entry and inelastic demand as two indicia of oligopoly). And while stable market shares over time may suggest in some factual contexts that the firms "eliminated competition[] among themselves," Posner, supra, at 79, it is also likely that conscious parallelism would be sufficient to maintain stable relative market shares in a stable market for a basic commodity protected from new entry. The district court specifically cited plaintiffs' expert's acknowledgment that the "nearly constant market shares are consistent with both cooperative and non-cooperative pricing behavior." See White, slip op. at 6.

Plaintiffs' seventh factor is that variations in price from region to region may indicate collusion. This may be true on some facts. See Posner, supra, at 87. But plaintiffs err in pointing to the difference in gas prices between Cape Cod and Martha's Vineyard. Even discounting the higher transportation costs of getting fuel to the island gas stations, we have already discussed a number of lawful reasons why the island stations are likely to be able to maintain the rest of the variation in price without agreement.

Plaintiffs' eighth purported plus factor is that Drake's employment of a consultant to lobby the Martha's Vineyard Commission to deny a petition for a new gas station on the island proves that Drake "was willing to act secretly to influence gas prices." Drake's unilateral retention of a consultant was a legitimate exercise of its right to petition. There is no claim that the consultant or the Commission acted illegally. In addition, this was entirely economically rational behavior, whatever the Commission decided. Indeed, these actions are well within those actions which the Supreme Court sought to protect from chilling effects. See Monsanto, 465 U.S. at 763, 104 S.Ct. 1464.

Other pieces of evidence require further examination and we consider them the strongest evidence plaintiffs have advanced. Plaintiffs produced some evidence that two of the defendants' principals communicated with one another and may have been untruthful about the communication. This fits with a different type of plus factor: "traditional" conspiracy evidence of the type that helps to distinguish between conscious parallelism and collusion and that is necessary to an inference of agreement. Flat Glass, 385 F.3d at 362; 6 Areeda & Hovenkamp, supra, ¶ 1434b, at 267. This need not be direct evidence; circumstantial evidence can suffice to establish an antitrust conspiracy. Monsanto, 465 U.S. at 764, 104 S.Ct. 1464.

Two witnesses recalled meetings held ten years earlier, in December 1999, with Jim Ahern of defendant Drake about contracting with Drake to be the wholesale gas supplier for a new gas station they were trying to open on the island. At the first meeting, the two described a ten-cent retail price discount for year-round Vine-yard residents that they proposed to offer in order to convince the Martha's Vineyard Commission to approve their permit. Ahern replied that he would not give year-round residents a discount.9 Then, referring to Ralph Packer, whose company R.M. Packer was the owner of Tisbury Shell, Ahern made the point that businesses on the Vineyard were also not inclined to give discounts, saying, "Your boy Ralph Packer, an island boy, is not cutting people any slack, and I'm not going to either." It is not clear from the record whether Ahern was speaking as a wholesaler or retailer in refusing to contemplate a discount, since both his company and Ralph Packer's company, R.M. Packer, played both wholesaler and retailer roles on Martha's Vineyard. R.M. Packer supplied other stations on the island that are not defendants. Drake supplied the two defendant stations owned by Paciello, but also was a wholesaler to a number of stations elsewhere on the East Coast.

At the second meeting, during a discussion between the two witnesses and Ahern about the discount, when the subject of Ralph Packer arose Ahern said, "You know we talk." Ahern then called on speakerphone a person whom Ahern said was Packer. It appeared to the witnesses that the call's recipient recognized Ahern's voice without Ahern identifying himself. We assume, in plaintiffs' favor, that the call recipient was Packer. Ahern casually stated to Packer that he was just checking in, and hung up after a friendly, general conversation of under five minutes. Ahern and Packer did not discuss the witnesses' desire to open a station, pricing or other business. One witness said he thought Ahern made the call "to demonstrate ... that it didn't really matter who we chose as a distributor, we'd have the same wholesale price." The witness also said that after this conversation, Ahern told them, "we all work together," and said, "I talk to Packer frequently."

Yet Ahern said in a 2009 deposition that he did not recall talking to Ralph Packer on the phone and only had one business meeting with him. Packer, in a 2008 deposition, said he met once with Ahern about wholesale supply about fifteen years earlier and had no phone conversations with him. In context, even if Ahern's and Packer's denials that they spoke over the phone were untrue, Ahern's actions and statements demonstrated at most a wholesaler's attempt to show his leverage over a potential wholesale customer, and are vague, ambiguous, and non-probative with regard to retail pricing practices.

There was more evidence about that meeting. One of the witnesses seeking permission to open a new station stated his view that if the new station were to offer gasoline at a lower price, there would be a chain reaction and the other stations would also lower their prices. In response to conjecture on what would happen if the two Paciello stations supplied by Ahern's company, Drake, started dropping their prices, Ahern replied, "if they start mucking around with prices one or two delivery trucks a week might not make it on the boat and they'll get the idea real quick." The witness "didn't really get the feeling... that it was a threat." This statement, in any event, was about unilateral action by Drake against one of its wholesale customers. It suggests competition between Drake's retail station and its wholesale customers' retail stations, not collusion among retailers.

Drake's and R.M. Packer's dual status as both wholesalers and retailers is thus relevant to understanding why the evidence from these meetings does not suffice to raise an inference of an agreement to fix prices in the retail market. Nothing forbids producers from selling in two different levels of the same market, here the wholesale and retail levels. Cf. Texaco Inc. v. Hasbrouck, 496 U.S. 543, 546-50, 110 S.Ct. 2535, 110 L.Ed.2d 492 (1990) (describing Texaco's arrangements selling gasoline to both distributors and retailers). Producer actions in the two levels cannot be conflated to produce an antitrust violation when there is no violation in either market alone. See Pac. Bell Tel. Co. v. Linkline Commc'ns, Inc., 555 U.S. 438, 129 S.Ct. 1109, 1121, 172 L.Ed.2d 836 (2009) (stating, "It is difficult enough for courts to identify and remedy an alleged anticompetitive practice at one level," and declining "simultaneously to police both the wholesale and retail" levels for price-squeezing under § 2 of the Sherman Act).

Plaintiffs have not argued that wholesaler-retailer relationships created vertical restraints on trade, such as minimum resale prices, affecting the retail market, and any such restraint if proven would not be illegal per se, but would be subject to analysis for whether it had anticompetitive effects on the market. See Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 892-94, 127 S.Ct. 2705, 168 L.Ed.2d 623 (2007). Nor have plaintiffs argued that Drake and R.M. Packer agreed to fix wholesale prices.

Plaintiffs also point to doubts about Packer's testimony that Tisbury Shell's profits were less than $100,000 a year and that there had not been any dividends for more than a decade from the R.M. Packer Company to Packer and his wife, the sole shareholders. Plaintiffs produced evidence that the R.M. Packer Company, in which Tisbury Shell accounted for about 40% of the revenues, had profits of more than $800,000 a year, and that Packer and his wife had received more than $950,000 in dividends from the company from 2003 to 2007. The apparent inaccuracies in Packer's deposition testimony are concerning, but the reason for them is unclear. At most there are inferences, which include Packer's confusion about the structure of his business and the lack of separate financial statements for Tisbury Shell's portion of the overall R.M. Packer company. The inaccuracies, though, more importantly, are not about whether there was an illegal agreement, but only about the amount of profits. Whatever their cause, these statements do not support any inference of conspiracy. Plaintiffs themselves concede that such "`pretext' standing alone is not sufficient to show joint action," but can only strengthen an inference of joint action that is otherwise in evidence. See DeLong Equip. Co. v. Washington Mills Abrasive Co., 887 F.2d 1499, 1514 (11th Cir.1989).

Plaintiffs have evidence of a highly favorable loan made by Drake to Paciello, owner of Edgartown Mobil and Depot Corner. Plaintiffs say that it follows from this that Paciello as retailer would agree to whatever Drake, wearing its retailer hat, wanted. The defendants say there was a legitimate business rationale. This was a loan from Drake as a wholesaler trying to keep a significant customer in business so that Drake's wholesale business would not diminish and would thrive. This also explains the favorable loan rates. In any event, it is not reasonable to infer from this that what Drake wanted was an illegal agreement to fix prices and that Paciello did so agree. Paciello might be rendered more pliable, but the loan evidence does not advance the likelihood of a tacit or express agreement.

The sum total of this evidence simply does not rise to the level Matsushita requires. Plaintiffs' ambiguous evidence is entirely consistent with permissible conscious parallelism. See Matsushita, 475 U.S. at 588, 106 S.Ct. 1348. Plaintiffs have failed to produce evidence that "tends to exclude the possibility of independent action." Monsanto, 465 U.S. at 768, 104 S.Ct. 1464.

Plaintiffs' final argument is that the district court improperly disregarded their expert's report, including his conclusion that defendants' "cost trends and coincident defendant pricing patterns are inconsistent with independent, non-cooperative behavior." The district court expressly adopted the expert's conclusions about the price and cost disparities between Martha's Vineyard and Cape Cod gas stations, and cited the expert's conclusions about the ambiguity of plaintiffs' market structure evidence. It is clear from the district court's opinion that it did not disregard the report, and the report does not undermine the conclusion that plaintiffs have failed to show that an agreement among the defendants is a more likely explanation for their pricing behaviors than bare conscious parallelism.

The expert acknowledged, "The results of my investigations are mixed," explaining that plaintiffs' economic evidence includes "instances that ... are not inconsistent with either cooperative or non-cooperative behavior" and other instances that are inconsistent with "independent, non-cooperative behavior." The report further acknowledged that defendants' decisions to increase prices as their costs declined, instead of undercutting other stations' prices, make no business sense "unless you know your rivals will follow," (emphasis added)—as they will, in conscious parallelism, in an interdependent oligopoly market.

As this court has said, "A firm in a concentrated industry typically has reason to decide (individually) to copy an industry leader. After all, a higher-than-leader's price might lead a customer to buy elsewhere, while a lower-than-leader's price might simply lead competitors to match the lower price, reducing profits for all. One does not need an agreement to bring about this kind of follow-the-leader effect in a concentrated industry." Clamp-All Corp., 851 F.2d at 484.

The antitrust claims fail.

III. Price-Gouging Claim

Massachusetts regulatory law prohibits selling gasoline at unconscionably high prices during market emergencies. See Mass. Gen. Laws ch. 93A § 2(a), (c); 940 Mass.Code Regs. 3.18. The district court entered summary judgment against plaintiffs' state law claims that the defendants engaged in price-gouging in the aftermath of Hurricanes Katrina and Rita in 2005.

The rule states,

Defendants have conceded for the purpose of their motions for summary judgment that a market emergency10 began on August 29, 2005, the day Hurricane Katrina made landfall in the United States, continued through the period following Hurricane Rita, which made landfall on September 24, and ended on December 1, 2005.11 The regulation defines neither "gross disparity" nor "immediately prior." Significantly, there have been no Massachusetts state court decisions interpreting the rule. Like the district court we write on a clean slate.12

Plaintiffs argue that the district court erred in rejecting their interpretation that a "gross disparity" can be proven from a change in profit as well as from a change in retail price. The district court used a "plain language" interpretation that because the regulation defines when "[a] price is unconscionably high," 940 Mass. Code Regs. 3.18(2), the analysis turns on disparities among prices at differing times and places. Absent such a showing about prices, the district court held, high profit margins cannot prove unconscionability of prices.

Plaintiffs argue that section 2 of the regulation provides only a nonexclusive method of proving that prices are unconscionably high, and that section 2(b) contemplates examination of gross margins (as a proxy for profit margins). They also argue that the regulation, whatever its language, must be interpreted to be consistent with the Federal Trade Commission's interpretations of different price-gouging definitions, because the Massachusetts state courts refer to certain FTC interpretations when interpreting Mass. Gen. Laws ch. 93A, the statute underlying the price-gouging regulation.13 See Ciardi v. F. Hoffmann-La Roche, Ltd., 436 Mass. 53, 762 N.E.2d 303, 309 (2002).

The facts about prices during this period are not disputed. Plaintiffs summarize their evidence as showing that from August 30, the day after Katrina made landfall, to the end of the emergency period, the absolute maximum increase in defendants' gross margin per gallon of regular gas ranged from 36 to 51 cents, representing 38% to 68% increases. Using less volatile monthly averages, plaintiffs' Exhibit 13 shows maximum increases in gross margins during the market emergency of 25 cents or 38% at R.M. Packer's Tisbury Shell, 35 cents or 69% at Drake's Xtra-Mart Citgo, 31 cents or 51% at Paciello's Edgartown Mobil, and 31 cents or 54% at Depot Corner. Tisbury Shell's maximum margin was in November; the other three stations' were in October.

Focusing on absolute changes in price per gallon from the week before to the week following Hurricane Katrina's August 29 landfall, prices rose 20 cents at Tisbury Shell, 48 cents at XtraMart Citgo, 42 cents at Edgartown Mobil, and 42 cents at Depot Corner. Prices continued to rise, to a maximum increase in early September of 37 cents at Tisbury Shell14 and 60 cents at XtraMart Citgo,15 Edgartown Mobil,16 and Depot Corner.17 During the week beginning on September 27, just after Hurricane Rita made landfall on September 24, prices were still above their August 22-28 pre-Katrina levels by 32 cents at Tisbury Shell, 42 cents at XtraMart Citgo, and 35 cents at both Edgartown Mobil and Depot Corner.

The language of the price-gouging regulation does not reach gross disparities in price alone. The regulation is concerned with increases in both price and cost, the two factors that determine gross margin. We need not address the separate issue of whether a gross disparity between pre-emergency and post-emergency gross margins might make out a claim of price-gouging where an increase in absolute price does not in itself appear unconscionable. See, e.g., People ex rel. Spitzer v. My Serv. Ctr., Inc., No. 06-21157, 2007 WL 102463, at *2 (N.Y.Sup.Ct. Jan. 17, 2007) (using increase in gross margin from 67 to 99 cents to illustrate that price increase of about 32 cents was price-gouging under New York law); People ex rel. Spitzer v. Wever Petroleum, Inc., 14 Misc.3d 491, 827 N.Y.S.2d 813, 816 (N.Y.Sup.Ct.2006) (finding increase of 60 cents in gross margin made increase of 87 cents in price "unconscionably excessive" under New York law).

The rule encompasses price and margin increases in relation to one another. Dramatic changes in gross margin might illustrate that a price increase is a "gross disparity" in price because it reflects price increases unexplained by cost increases. But nothing in the regulation suggests that increases in gross margin alone, in the absence of any price increase and simultaneous with declining retail prices, can support a price-gouging claim.

While there is no specific history available as to the Massachusetts price-gouging rule, such rules are generally designed to protect consumers from acute and unconscionable increases in the prices they must pay for basic consumer goods during times of market emergency, not to mandate that retailers decrease their prices as quickly as their costs decline after the most acute crisis in supply of the good has passed. See, e.g., Ark.Code Ann. § 4-88-301 (declaring legislative purpose to prevent "excessive and unjustified increases" in prices); Cal.Penal Code § 396 ("It is the intent of the Legislature ... to protect citizens from excessive and unjustified increases in the prices charged. ..."). These are not regulations meant to give the government control over the setting of petroleum product prices.

The facts show that while defendants' average weekly prices were increasing, during the time periods of August 30-September 5 and September 6-12, their gross margins were generally rising only very moderately, since their costs were climbing as well.18 Average margins rose at Tisbury Shell from 74 cents during the August 22-28 period to 78 cents from August 30-September 5, before dropping back down to 61 cents the following week. During the same time periods, margins at XtraMart rose from 53 cents to 87 cents before falling back to 68 cents, margins at Edgartown Mobil rose from 66 cents to 68 cents and then to 76 cents, and margins at Depot Corner rose from 62 cents to 64 cents and then to 72 cents.

There is marked volatility in the margins, best shown by the swing in average gross margin at XtraMart Citgo from 53 cents during the week before Katrina to 87 cents during the first week after, but dropping back to 68 cents one week later as its costs continued to rise. This volatility resulted from mismatches between when the stations raised their prices and when they had to pay higher costs at wholesale.19 At XtraMart, for example, prices rose from $3.50 on September 1 to $3.70 a gallon on September 2, when costs were $2.47 a gallon, but there was no further increase in price at XtraMart after costs rose to $3.15 a gallon on September 5. And the initial twenty-cent price increase to $3.70 a gallon itself came two days after a short-lived but sharp increase in XtraMart's costs from $2.56 to $2.93 per gallon, before costs dropped to $2.47 on September 2. Unless the resulting prices are "unconscionably high," the price-gouging rule does not prohibit retailers from raising their prices in reasonable anticipation of future increases in costs, or after the fact in response to actual recent increases even if costs have dropped back down again.

Defendants' margins hit their highest levels after their retail prices began to decline. It appears that the stations' costs dropped precipitously beginning in mid-September, but that they dropped their prices at a much slower rate. The stations' average weekly gross margins thus began a general rise from the week of September 13 well into October and November. But no station raised its price after September 7, and all four had dropped from their highest price by September 20 at the very latest.

Neither the absolute increases in price nor the increases in gross margins show any "gross disparity" in price. As we have mentioned, no Massachusetts law defines "gross disparity" for the purposes of the price-gouging regulation. By analogy, however, in the context of unconscionable contracts, a Massachusetts case does refer to a "gross disparity" as requiring "gross inadequacy of consideration." Waters v. Min Ltd., 412 Mass. 64, 587 N.E.2d 231, 234 (1992) (quoting 1 A. Corbin, Contracts § 128, at 551 (1963 & Supp. 1991)). That court said the disparity must "lead[] inevitably to the felt conclusion that knowing advantage was taken" of consumers. Id. at 233 (quoting Jones v. Star Credit Corp., 59 Misc.2d 189, 298 N.Y.S.2d 264 (N.Y.Sup.Ct.1969)). Those standards have not been met here.

The Federal Trade Commission report on which plaintiffs rely would not lead us to a different result, and so we do not decide what, if any, deference the Massachusetts Supreme Judicial Court would give it. See Federal Trade Commission, Investigation of Gasoline Price Manipulation and Post-Katrina Gasoline Price Increases (2006). Plaintiffs cite the FTC's statement that an increase in average margin of more than five cents is "a price increase that was not substantially explained by increased costs." Id. at 151. But they incorrectly state that the FTC equated this five-cent margin increase with price-gouging. In fact, the FTC Report required an additional conclusion—that a retailer's absolute price increase exceeded the national average increase of thirty-five cents, as well as the average increase in the station's local area, by at least five cents—before it was considered to have engaged in price-gouging. Id. at 152. Under plaintiffs' expert's definition of Cape Cod as the relevant "trade area" under the Massachusetts price-gouging rule, none of defendants' gas stations were price-gouging under the FTC's methodology: average gas prices on the Cape increased by 53 cents from August to September, while defendants' average gas prices increased between 42 and 54 cents.20

Plaintiffs have not shown a "gross disparity" in prices under the state price-gouging rule, even taking into account defendants' gross margins during the period of price increases.

The only question before us is whether the supracompetitive prices charged by defendants on Martha's Vineyard are a result of illegal actions in violation of federal antitrust laws or state anti-price-gouging rules. Plaintiffs have failed to meet the legal standards for proof of those violations. The judgment of the district court is affirmed.