*969A two-letter conjunction and a two-word phrase decide this case. At stake are hundreds of millions of dollars. Congress, acting shortly after the economy began to stabilize from the financial crisis that began a decade earlier, passed the Dodd-Frank Wall Street Reform and Consumer Protection Act, Pub. L. No. 111-203, 124 Stat. 1376 (2010), which amended the Commodity Exchange Act (CEA) to expand the Commodity Future Trading Commission's (CFTC) enforcement authority. This case is about the extent of those powers.
Monex Credit Company, one of the defendants and appellees, argues that the CFTC went too far when it filed this $290 million lawsuit for alleged fraud in precious metals sales. According to Monex, Dodd-Frank extended the CFTC's power only to fraud-based manipulation claims, so stand-alone fraud claims-without allegations of manipulation-fail as a matter of law.
Not only that, Monex argues, but Dodd-Frank also immunizes Monex from the CFTC's claims that it ran an unregistered, off-exchange trading platform. The CEA's registration provisions do not apply to retail commodities dealers who "actual[ly] deliver[ ]" the commodities to customers within twenty-eight days. See 7 U.S.C. § 2(c)(2)(D)(ii)(III)(aa). Monex insists that it falls within this exception.
On both fronts, the district court agreed with Monex and dismissed the CFTC's complaint for failure to state a claim under Civil Rule 12(b)(6). We REVERSE and REMAND .
Background
The facts come from the CFTC's complaint, which, at this stage, we must accept as true. See Syed v. M-I, LLC , 853 F.3d 492, 499 (9th Cir. 2017).
Monex and the Atlas Program
California-based Monex has been a major player in the precious metal markets for decades. It sells gold, silver, platinum, and palladium to investors who have a variety of buying options, but here we focus on what Monex calls its "Atlas Program." Through Atlas, investors can purchase commodities on "margin." Also known as "leverage," the concept is simple: A customer buys precious metals by paying only a portion of the full price. The remaining amount is financed through Monex.
Once a customer opens an account, she may take open positions in precious metals. But the trading occurs "off exchange"-that is, it does not happen on a regulated exchange or board of trade. Instead, *970Monex controls the platform, acts as the counterparty to every transaction, and sets the price for every trade.
Since mid-2011, Monex has made more than 140,000 trades for more than 12,000 Atlas accounts, each of which requires margin of 22-25% of the account's total value. A customer who deposits $25,000 in Atlas as margin can open positions valued at $100,000; she owes the additional $75,000 to Monex. Over time, the account's value changes-it goes up and down-as markets do. The difference between the account's total value and the amount the customer still owes to Monex is the account's "equity." And if that difference falls below a certain threshold, Monex can issue a "margin call"-it can require customers to immediately deposit more money into the accounts to increase the equity. Monex can do so at any time, and it can change margin requirements whenever it wants.
Monex also retains sole discretion to liquidate trading positions without notice to the customer if equity drops too low, and it controls the price for every trade. Price spreads-the difference between the bid price and ask price-are 3% and generate much of the program's revenue. Commissions and fees make up the rest, and that money comes directly out of customer accounts' equity. Over the last eight years, Monex has made margin calls in more than 3,000 Atlas accounts and has force-liquidated at least 1,850.
Atlas investors can make either "short" or "long" trades. Short trades bet on metal prices going down, and long up. Monex allows investors to place "stop" or "limit" orders to manage their trading positions. About a quarter of trading positions in leveraged Atlas accounts open and close within two weeks.
Customers must sign the Atlas account agreement, which gives Monex control over the metals. Monex does not hand over any metals, and customers never possess or control any physical commodity. Instead, Monex stores the metals in depositories with which Monex has contractual relationships. Monex retains exclusive authority to direct the depository on how to handle the metals; investors and the depositories have no contractual relationship with each other. Customers can get their hands on the metals only by making full payment, requesting specific delivery of metals, and having the metals shipped to themselves, a pick-up location, or an agent.
This structure applies to both long and short positions. For a long position, Monex retains the right to close out the position at any time in its sole discretion and at a price Monex chooses. Metal remains in the depository, but Monex claims to transfer ownership of the metals to the customer. The same is true for short positions, except that instead of transferring ownership, Monex loans the customer metals that the customer immediately sells back to Monex. According to the CFTC, Monex simply makes a "book entry" when customers make trades-nothing more.
The Commodity Exchange Act and Dodd-Frank
The CFTC regulates commodity futures markets under the CEA. See 7 U.S.C. §§ 1 et seq . Part of the CEA's purpose is "to protect all market participants from fraudulent or other abusive sales practices and misuses of customer assets." Id. § 5(b). The CEA requires that futures be traded on regulated exchanges. Id . § 6(a)(1). Brokers must register with the CFTC. Id . § 6d(a)(1). The CEA further protects against conflicts of interest and market abuse. Id . §§ 6d(c), 7(d). And the statute prohibits fraud. Id . § 6b(a)(2).
*971Originally, the CEA did not apply to retail commodity transactions because they were not futures contracts. See CFTC v. Zelener , 373 F.3d 861 (7th Cir. 2004). As Zelener recognized, the CEA applied only to futures contracts, even though other types of sales-such as leveraged retail commodity sales-can have similar economic effects. Id . at 866-67.
This changed in 2010 when Congress, acting in the wake of financial turmoil, passed Dodd-Frank-part of which amended the CEA. See Dodd-Frank Wall Street Reform and Consumer Protection Act of 2010, Pub. L. No. 111-203, 124 Stat. 1376 (2010). Congress extended the CEA to commodity transactions offered "on a leveraged or margined basis, or financed by the offeror" "as if" they were futures trades. See 7 U.S.C. § 2(c)(2)(D)(iii). But Congress carved out an exception: The CEA would not apply to leveraged retail commodity sales that resulted "in actual delivery within 28 days." Id . § 2(c)(2)(D)(ii)(III)(aa).
Congress also amended the CEA by prohibiting the use of "any manipulative or deceptive device or contrivance" in market transactions. CEA § 6(c)(1). This language mirrored § 10(b) of the Securities and Exchange Act, and, as did § 10(b), authorized the governing agency to promulgate rules implementing the statute and bring civil enforcement actions. See 7 U.S.C. §§ 9(1), 13a-1(a) ; 15 U.S.C. § 78j(b).
Monex's Alleged Scheme and This Lawsuit
The CFTC contends that Atlas is a scheme that has violated the CEA since at least July 2011. Monex tells its customers that leveraged precious metals trading is "a safe, secure and profitable way for retail customers to invest" when, in fact, the program requires that many customers lose money. What's more, the CFTC alleges, Atlas is designed so that when customers lose, Monex gains: Because Monex is the counterparty for each Atlas transaction, Monex benefits from large price spreads at the customer's expense. Sales representatives, too, have an incentive to push the program: Monex pays salespeople with "commissions and bonuses tied directly to the number of Atlas accounts they open" and the number of transactions completed; account performance is not a factor in compensation. So Monex engages in "high-pressure sales tactics," cajoling potential customers into buying leveraged precious metals while it "misrepresent[s] the likelihood of profit" and "systematically downplay[s] the risks" to ensure customers invest in Atlas, inevitably leading to customer losses.
The complaint alleges deep and broad losses to about 90% of all leveraged Atlas accounts-totaling some $290 million. In some cases, individual losses were extreme: some customers lost hundreds of thousands of dollars, and many others suffered five-figure losses. New investors never learned about those losses because Monex never told them. Instead, Monex promised that precious metals are safe and "will always have value," so a customer cannot lose her investment.
The CFTC filed this lawsuit seeking an injunction and restitution against Monex Deposit Company, Monex Credit Company, Newport Services Corporation, Louis Carabini, and Michael Carabini (Monex). The CFTC contends that Atlas is an illegal and unregistered leveraged retail commodity transaction market. The CFTC filed four counts, alleging violations of:
(1) CEA § 4(a), 7 U.S.C. § 6(a), for engaging in off-exchange transactions;
(2) CEA § 4b(a)(2)(A) and (C), 7 U.S.C. § 6b(a)(2)(A) and (C), for fraud;
*972(3) CEA § 6(c)(1), 7 U.S.C. § 9(1), 17 CFR § 180.1(a)(1)-(3), for fraud; and
(4) CEA § 4d, 7 U.S.C. § 6d(a)(1), for failing to register.
The CFTC filed this lawsuit in the Northern District of Illinois in September 2017. The same day, the CFTC moved for a preliminary injunction. A month later, Monex filed a motion to dismiss for failure to state a claim under Civil Rule 12(b)(6). The Illinois district court transferred the case to the Central District of California three weeks later.
The District Court Dismisses the CFTC's Complaint
The district court granted Monex's motion to dismiss, denied as moot the motion for preliminary injunction, and gave the CFTC thirty days to amend its complaint as to Count III, the CEA § 6(c)(1) fraud claim. The CFTC declined the invitation to amend and asked the court to enter judgment, which it did.
The district court determined that Counts I, II, and IV failed because Monex fit within the actual delivery exception. 7 U.S.C. § 2(c)(2)(D)(ii)(III)(aa). The district court dismissed Count III because § 6 allows the CFTC to bring only fraud-based manipulation claims-not stand-alone fraud cases. In short, the district court held that "any manipulative or deceptive device" in § 6(c)(1) requires manipulative and fraudulent behavior. And because the CFTC alleged only fraud-and not manipulation-Count III failed as a matter of law. This appeal followed.
Standard of Review
In reviewing a Civil Rule 12(b)(6) dismissal, we give no deference to the district court. Soltysik v. Padilla , 910 F.3d 438, 444 (9th Cir. 2018). This de novo review consists of two steps. First, we identify all the factual allegations in the complaint and accept them as true; legal conclusions are set aside. Ashcroft v. Iqbal , 556 U.S. 662, 678-79, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009). Second, reading all the allegations in the light most favorable to the non-moving party, we ask whether the facts state a claim for relief. Id. ; see Fed. R. Civ. P. 8(a). To survive, the claim must be plausible . Iqbal , 556 U.S. at 678, 129 S.Ct. 1937. That is, it must rise "above the speculative level." Bell Atl. Corp. v. Twombly , 550 U.S. 544, 555-56, 127 S.Ct. 1955, 167 L.Ed.2d 929 (2007). Claims move beyond speculation when the allegations "allow[ ] the court to draw the reasonable inference that the defendant is liable for the misconduct alleged." Iqbal , 556 U.S. at 678, 129 S.Ct. 1937. This is a "context-specific task that requires the reviewing court to draw on its judicial experience and common sense." Id . at 679, 129 S.Ct. 1937.
For claims of fraud, we require additional specificity: who, what, when, where, and how. See Fed. R. Civ. P. 9(b) ; Vess v. Ciba-Geigy Corp., USA , 317 F.3d 1097, 1106 (9th Cir. 2003).
Discussion
A. The Actual Delivery Exception
We must first determine whether the actual delivery exception is an element of a CEA claim or an affirmative defense. This distinction is important because Rule 8 does not require plaintiffs to plead around affirmative defenses. See Jones v. Bock , 549 U.S. 199, 216, 127 S.Ct. 910, 166 L.Ed.2d 798 (2007). And "[o]rdinarily, affirmative defenses ... may not be raised on a motion to dismiss." Lusnak v. Bank of Am., N.A. , 883 F.3d 1185, 1194 n.6 (9th Cir. 2018).
The Eleventh Circuit has ruled that the actual delivery exception "is an affirmative defense on which the commodities *973trader bears the burden of proof." CFTC v. S. Trust Metals, Inc ., 894 F.3d 1313, 1324-25 (11th Cir. 2018). We agree. Placing the burden on the defendant is, after all, the "general rule where [the defendant] claims the benefits of an exception to the prohibition of a statute." United States v. First City Nat'l Bank of Houston , 386 U.S. 361, 366, 87 S.Ct. 1088, 18 L.Ed.2d 151 (1967). And this "longstanding convention is part of the backdrop against which Congress writes laws," so courts must "respect it unless we have compelling reasons to think that Congress meant to put the burden of persuasion on the other side." Meacham v. Knolls Atomic Power Lab. , 554 U.S. 84, 91-92, 128 S.Ct. 2395, 171 L.Ed.2d 283 (2008).
Nevertheless, we can consider an affirmative defense on a motion to dismiss when there is "some obvious bar to securing relief on the face of the complaint." ASARCO, LLC v. Union Pac. R.R. Co. , 765 F.3d 999, 1004 (9th Cir. 2014). In other words, dismissal based on an affirmative defense is permitted when the complaint establishes the defense. See Sams v. Yahoo! Inc. , 713 F.3d 1175, 1179 (9th Cir. 2013). To determine whether Atlas, described in the CFTC's complaint, includes "actual delivery," we must identify the meaning of that statutory term.
Under CEA §§ 2(c)(2)(D)(i) and (iii), any "agreement, contract, or transaction in any commodity that is entered into ... on a leveraged or margined basis" is subject to " sections 6(a), 6(b), and 6b" of the CEA "as if the agreement, contract or transaction was a contract of sale of a commodity for future delivery." 7 U.S.C. §§ 2(c)(2)(D)(i) and (iii). But not all sales; the adjacent section excludes "a contract of sale that results in actual delivery within 28 days." Id. § 2(c)(2)(D)(ii)(III)(aa).
The statute does not define "actual delivery," and undefined terms receive their ordinary meaning. See Taniguchi v. Kan Pac. Saipan, Ltd , 566 U.S. 560, 566, 132 S.Ct. 1997, 182 L.Ed.2d 903 (2012). "Delivery" means "[t]he formal act of voluntarily transferring something; esp. the act of bringing goods, letters, etc. to a particular person or place." Black's Law Dictionary (9th ed. 2009). Black's defines "actual" as "[e]xisting in fact; real." Id. "Actual delivery" is the "act of giving real and immediate possession to the buyer or the buyer's agent." Id . By contrast, "constructive delivery" denotes "[a]n act that amounts to transfer of title by operation of law when actual transfer is impractical or impossible." Id .
The Eleventh Circuit adopted these definitions in CFTC v. Hunter Wise Commodities, LLC , 749 F.3d 967 (11th Cir. 2014), where it held that a seller failed to actually deliver commodities when it "did not possess or control an inventory of metal from which it could deliver to retail customers." Id . at 980. The court did "not define the precise boundaries of 'actual delivery,' " but it held that "[d]elivery must be actual ." Id . at 979 (emphasis in original). "If 'actual delivery' means anything, it means something other than simply 'delivery,' for we must attach meaning to Congress's use of the modifier 'actual.' " Id . The defendant in Hunter Wise could not actually deliver anything because it did not have the commodities.
According to Monex, Hunter Wise tells us that the actual delivery exception applies only when the commodities do not in fact exist. Monex argues that it makes actual delivery "because the metals exist in fact and, upon sale, are voluntarily delivered to independent depositories for the buyer's benefit." Appellee Br. at 10-11. Monex, unlike the defendant in Hunter Wise , has the underlying commodities-they actually exist. So, Monex argues, *974Hunter Wise does not apply, and Atlas fits the exception.
Hunter Wise is not so limited. That court first held that "actual delivery" means giving "real and immediate possession to the buyer or buyer's agent." Hunter Wise , 749 F.3d at 979 (quoting Black's Law Dictionary 494 (9th ed. 2009)). The seller in Hunter Wise did not give the buyer possession of the commodities because it did not possess any in the first instance. Id. Without inventory, the seller could not actually deliver anything. Id . But "actual" in the statute modifies delivery , not existence. See id . Of course, as Hunter Wise recognizes, existence is a prerequisite to delivery-one cannot deliver that which does not exist. But the fact that the commodity's existence is necessary to comply with the exception does not mean existence is sufficient to fit the exception. If Congress wanted only to ensure enough inventory it could have said so. It did not; it required "actual delivery."
Thus, the plain language tells us that actual delivery requires at least some meaningful degree of possession or control by the customer. It is possible for this exception to be satisfied when the commodity sits in a third-party depository, but not when, as here, metals are in the broker's chosen depository, never exchange hands, and are subject to the broker's exclusive control, and customers have no substantial, non-contingent interests.
This interpretation is confirmed by the broader statutory context. See Abramski v. United States , 573 U.S. 169, 179, 134 S.Ct. 2259, 189 L.Ed.2d 262 (2014). Dodd-Frank expanded the CEA to close the so-called Zelener loophole, which allowed companies to offer commodity sales on margin without regulation, because these transactions mimic conventional futures trades long regulated by the CFTC. See Zelener , 373 F.3d at 866. On the other hand, sales where customers obtain meaningful control or possession of commodities, i.e., when actual delivery occurs, do not mimic futures trading and are therefore exempt from registration and related CEA requirements.
Monex argues that in the context of a provision regulating leveraged commodity sales, it would make little sense for "actual delivery" to turn on possession or control, because such a reading would clash with "margin," which means "[c]ash or collateral required to be paid to a securities broker by an investor to protect the broker against losses from securities bought on credit." Black's Law Dictionary (9th ed. 2009). Because the very meaning of the word "margin" requires that the buyer deposit collateral with the seller, actual delivery must mean something other than transferring possession or control to the buyer. Otherwise, Monex argues, margin would mean nothing.
Yet, even if the commodity serves as collateral, there is no reason why the buyer cannot control it. In many financing contexts, some degree of buyer possession or control is commonplace. While permitting customers to obtain significant control over or possession of metals might be practically difficult here, that fact does not displace the statute's plain meaning.
If we had any lingering doubt about the statute's plain meaning, resort to conventional canons of interpretation would further support our conclusion. First, the CEA uses "delivery" in § 1a(27), which we have said "cannot be satisfied by the simple device of a transfer of title." CFTC v. Noble Metals Int'l, Inc. , 67 F.3d 766, 773 (9th Cir. 1995). And because we assume that "Congress means the same words in the same statute to mean the same thing," actual delivery must require more than simple title transfer.
*975Texas Dept. of Housing & Cmty. Affairs v. Inclusive Cmtys. Project, Inc. , --- U.S. ----, 135 S. Ct. 2507, 2535, 192 L.Ed.2d 514 (2015). Second, our interpretation presents no ineffectiveness or surplusage problems because it does not, as the district court believed, mean that "every financed transaction would violate Dodd-Frank," thus "eliminat[ing] the Actual Delivery Exception from the CEA." 311 F. Supp. 3d 1173, 1181 (C.D. Cal. 2018) (quoting CFTC v. Worth Grp., Inc ., No. 13-80796-CIV, 2014 WL 11350233, at *2 (S.D. Fla. Oct. 27, 2014) ). The CFTC does not present a bare-bones complaint. It includes detailed and specific factual allegations. All we say today is that those allegations, taken as true, do not establish actual delivery.
Finally, even if the statute were ambiguous, we would find the CFTC's interpretive guidance persuasive. Retail Commodity Transactions Under CEA , 78 Fed. Reg. 52,426 (Aug. 23, 2013) ; see Skidmore v. Swift & Co. , 323 U.S. 134, 65 S.Ct. 161, 89 L.Ed. 124 (1944). There, the CFTC stated it would employ a "functional approach" that considers "[o]wnership, possession, title, and physical location of the commodity purchased or sold." 78 Fed. Reg. at 52,428. Other factors included "the nature of the relationship between the buyer, seller, and possessor of the commodity," and the "manner in which the purchase or sale is recorded and completed." Id .
Monex insists that Atlas matches the second illustrative example of actual delivery set forth in the guidance: physical transfer of all purchased commodities into an independent depository plus transfer of title to the buyer. Id. However, these steps constitute actual delivery only if they are "not simply a sham." Id. The CFTC engages in a "careful consideration" of the relevant functional factors (listed above) to determine if the exception is indeed applicable. Here, customers have no contractual rights to the metal; Monex, not customers, has a relationship with depositories; Monex maintains total control over accounts and can liquidate at any time in its own discretion; and the entire transaction is merely a book entry. This amounts to sham delivery, not actual delivery.
To recap, "actual delivery" unambiguously requires the transfer of some degree of possession or control. Other interpretive tools, including the CFTC's guidance, reinforce this conclusion. Monex challenges the CFTC's characterization of its delivery scheme, but, at the 12(b)(6) stage, we ignore such factual disputes and accept as true allegations in the complaint. Because this affirmative defense does not, on the face of the complaint, bar the CFTC from relief on Counts I, II, and IV, the district court erred in dismissing those claims.
B. Manipulative or Deceptive
In Count III, the CFTC alleges that Monex violated CEA § 6(c)(1), 7 U.S.C. § 9(1), and 17 C.F.R. § 180.1 by fraudulently deceiving its customers. There is no allegation that Monex manipulated the market, so we must decide whether § 6(c)(1) covers fraud claims in the absence of manipulation. The text:
It shall be unlawful for any person, directly or indirectly, to use or employ, or attempt to use or employ, in connection with any swap, or a contract of sale of any commodity in interstate commerce, or for future delivery on or subject to the rules of any registered entity, any manipulative or deceptive device or contrivance, in contravention of such rules and regulations as the Commission shall promulgate.
7 U.S.C. § 9(1).
The crucial question is whether "any manipulative or deceptive device" allows stand-alone fraud claims or requires fraud-based manipulation. The district court determined *976that the statute unambiguously requires "both manipulative and deceptive conduct, not one or the other." Or, another way to say it, the district court held that "or" really meant "and." We disagree.
When the word "or" joins two terms, we apply a disjunctive reading. See, e.g. , United States v. Woods , 571 U.S. 31, 45-46, 134 S.Ct. 557, 187 L.Ed.2d 472 (2013). When Congress places "or" between two words, we assume that Congress intended the two terms as alternatives. See Scalia & Garner, Reading Law, § 12 at 116 (2012). While there are exceptions, this is not an instance where a disjunctive meaning would produce absurd results and statutory context compels us to treat "or" as if it were "and." See De Sylva v. Ballentine , 351 U.S. 570, 573, 76 S.Ct. 974, 100 L.Ed. 1415 (1956) ; United States v. Bonilla-Montenegro , 331 F.3d 1047, 1051 (9th Cir. 2003) ("a statute's use of disjunctive or conjunctive language is not always determinative"). We conclude that § 6(c)(1) 's language is unambiguous. Authorizing claims against "[m]anipulative or deceptive" conduct means what it says: the CFTC may sue for fraudulently deceptive activity, regardless of whether it was also manipulative.
Again, if we had any doubt, see Conn. Nat'l Bank v. Germain , 503 U.S. 249, 253-54, 112 S.Ct. 1146, 117 L.Ed.2d 391 (1992), other interpretive tools support our conclusion. This CEA provision is a mirror image of § 10(b) of the Securities Exchange Act, which the Supreme Court has interpreted as a "catch-all clause to prevent fraudulent practices," Chiarella v. United States , 445 U.S. 222, 226, 100 S.Ct. 1108, 63 L.Ed.2d 348 (1980), that authorizes fraud-only claims, see SEC v. Zandford , 535 U.S. 813, 822-25, 122 S.Ct. 1899, 153 L.Ed.2d 1 (2002). We presume that by copying § 10(b)'s language and pasting it in the CEA, Congress adopted § 10(b)'s judicial interpretations as well. Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Dabit , 547 U.S. 71, 85-86, 126 S.Ct. 1503, 164 L.Ed.2d 179 (2006).
The canon against surplusage does not point to a different answer: § 6(c)(1) 's overlap with other provisions is minimal, and partial redundancy hardly justifies displacing otherwise clear text. See J.E.M. Ag Supply, Inc. v. Pioneer Hi-Bred Int'l, Inc. , 534 U.S. 124, 144, 122 S.Ct. 593, 151 L.Ed.2d 508 (2001). Nor does the fact that the applicable statutory headings mention only manipulation and not fraud. The full extent of a statutory provision rarely fits into its title, so headings are often under inclusive. See Lawson v. FMR LLC , 571 U.S. 429, 446, 134 S.Ct. 1158, 188 L.Ed.2d 158 (2014). Finally the CEA elsewhere references a "manipulative device or contrivance," see 7 U.S.C. § 25(a)(1)(D)(i), suggesting that Congress knew how to require market manipulation when it sought to do so. The inclusion of "deceptive" in § 6(c)(1) must have meaning.
Monex pulls two final arrows from its quiver. First, Monex argues that the CFTC's enforcement jurisdiction comes only from CEA § 2. Without an independent jurisdictional grant in § 2, Monex argues, the CFTC cannot bring a § 6(c)(1) fraud claim. In support, Monex cites CFTC v. White Pine Tr. Corp. , 574 F.3d 1219 (9th Cir. 2009), where we considered whether the CFTC had jurisdiction over certain foreign currency trades. There, we focused on CEA § 4c, which applies only to a "transaction involving any commodity regulated under this chapter." 7 U.S.C. § 6c(b). The question in White Pine was whether foreign currency trades were "regulated under this chapter." 574 F.3d at 1223. Section 2 of the CEA generally excludes foreign currency from regulation, see § 2(c)(1), but some foreign currency are covered, see § 2(c)(2). Reading §§ 4c, *9772(c)(1), and 2(c)(2) together, we held in White Pine that the specific trades in that case did not fall under the CFTC's jurisdiction because foreign currency trades were categorically excluded from the CEA under § 2(c)(1), unless they were trades specifically exempted from that exclusion under § 2(c)(2). The White Pine trades did not fall under § 2(c)(2), and thus were excluded under § 2(c)(1). Id.
As the district court noted, retail commodity transactions are not addressed in § 2(c)(1) 's general exclusion. Thus, there is no need for a specific jurisdictional grant to overcome the general exclusion, as was required in White Pine . Instead, the retail commodity provision merely describes the types of transactions to which other CEA sections-§§ 4(a), 4(b), and 4b-apply. In other words, § 2(c)(2)(D) -the retail commodity provision-clarifies the interplay between margined commodity sales and other sections that apply to future contracts. This is necessary because §§ 4(a), 4(b), and 4b applied only to futures trades, until § 2(c)(2)(D) confirmed that those sections also apply to leveraged commodity sales.
No such clarification is needed with § 6(c)(1) because the section applies to "any ... contract of sale of any commodity in interstate commerce." And in those sales, § 6(c)(1) outlaws the use of any manipulative or deceptive device. Later, the CEA clarifies that "[w]henever it shall appear to the Commission that any registered entity or other person has" violated "any provision of this chapter ... the Commission may bring an action in the proper district court of the United States." 7 U.S.C. § 13a-1(a). When someone violates § 6(c)(1), the CFTC can bring an enforcement action.
Finally, Defendants argue that if § 6(c)(1) means what the CFTC says, then the statute applies not only to margined commodity sales, but to ordinary retail cash commodity sales, too. As Monex tells it, this would mean that even everyday grocery sales would be subject to the CFTC's enforcement power. See Appellee Br. at 35. This, Monex argues, cannot be the case because such an "explosive increase of an agency's ... authority" requires a clear statement from Congress. Id . at 53. And "Congress ... does not alter the fundamental details of a regulatory scheme in vague terms or ancillary provision-it does not, one might say, hide elephants in mouseholes." Whitman v. Am. Trucking Ass'ns , 531 U.S. 457, 468, 121 S.Ct. 903, 149 L.Ed.2d 1 (2001).
In the first place, it is not clear that this amounts to an elephant in a mousehole. By its terms, § 6(c)(1) applies broadly to commodities in interstate commerce. More important, this case does not involve retail cash commodity sales. This case involves only margined commodity sales. And even Monex admits that § 6(c)(1) applies to at least some margined commodity sales-those that involve fraud-based manipulation. The question we address is only whether § 6(c)(1) also applies to stand-alone fraud claims in the sale of leveraged commodities. Whether the statute extends to non-leveraged sales is not before us.
Conclusion
In bill drafting, as in life, little things often make big differences. Here, three words stand between dismissal and discovery. Although Monex contends that no fraud occurred, we must, at this point, accept as true the CFTC's well-pleaded complaint to the contrary. And because the CFTC's claims are plausible, this lawsuit should continue.
REVERSED and REMANDED for further proceedings consistent with this decision. 1
*978--------