American Equity Investment Life Insurance v. Securities & Exchange Commission

United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT Argued May 8, 2009 Decided July 21, 2009 No. 09-1021 AMERICAN EQUITY INVESTMENT LIFE INSURANCE COMPANY, ET AL., PETITIONERS v. SECURITIES AND EXCHANGE COMMISSION, RESPONDENT Consolidated with 09-1056 On Petitions for Review of an Order of the Securities & Exchange Commission Eugene Scalia argued the cause for petitioners American Equity Investment Life Insurance Company, et al. With him on the briefs were Barry Goldsmith and Daniel J. Davis. Rodney F. Page argued the cause and filed the briefs for petitioner National Association of Insurance Commissioners. Julius A. Rousseau entered an appearance. 2 Kenneth W. Sukhia was on the brief for amici curiae Phillip Roy Financial Services, LLC and Phillip R. Wasserman in support of petitioners. James F. Jorden, Frank G. Burt and Gary O. Cohen were on the brief for amicus curiae Allianz Life Insurance Company of North America in support of petitioners. Michael A. Conley, Deputy Solicitor, Securities & Exchange Commission, argued the cause for respondent. With him on the brief were David M. Becker, General Counsel, Jacob H. Stillman, Solicitor, and Dominick V. Freda and William K. Shirey, Senior Counsel. Deborah M. Zuckerman, Rex Staples, Stephen Hall, and Michael Lacek were on the brief for amici curiae AARP, et al. in support of respondent. Michael R. Schuster entered an appearance. Before: SENTELLE, Chief Judge, and GINSBURG and ROGERS, Circuit Judges. Opinion for the Court filed by Chief Judge SENTELLE. SENTELLE, Chief Judge: The Securities Act of 1933, 15 U.S.C. §§ 77a et seq. (the Act), exempts from federal regulation annuity contracts issued by a corporation subject to regulation by state insurance laws. Petitioners seek review of a rule promulgated by the Securities and Exchange Commission (SEC or Commission) stating that fixed indexed annuities (FIAs) are not annuity contracts within the meaning of the Act. As a result of this new rule, FIAs are subject to the full panoply of requirements set forth by the Act, instead of being subject solely to state insurance laws. Petitioners argue that the Commission unreasonably interpreted the term “annuity contract” not to 3 include FIAs. Petitioners also assert that the SEC failed to fulfill its statutory responsibility under the Act to consider the effect of the new rule on efficiency, competition, and capital formation. Because we hold that the SEC’s interpretation of “annuity contract” is reasonable under Chevron, we deny the petitions with respect to this issue. We grant the petitions, however, with respect to petitioners’ alternate ground that the SEC failed to properly consider the effect of the rule upon efficiency, competition, and capital formation. Accordingly, we remand the rule for reconsideration. I. BACKGROUND A. The Securities Act of 1933 governs the offer or sale of any security through interstate commerce. The Act defines the term “security” as including any “investment contract.” 15 U.S.C. § 77b(a)(1); SEC v. Variable Annuity Life Ins. Co. of Am. (VALIC), 359 U.S. 65, 67-68 (1959). Section 3(a)(8) of the Act, however, provides an exemption under the Act for an “annuity contract” or “optional annuity contract” subject to state insurance laws. 15 U.S.C. § 77c(a)(8). A traditional fixed annuity is a contract issued by a life insurance company, under which the purchaser makes a series of premium payments to the insurer in exchange for a series of periodic payments from the insurer to the purchaser at agreed upon later dates. In a fixed annuity, the insurance company guarantees that the purchaser will earn a minimum rate of interest over time. Fixed annuities are subject to state insurance law regulation, and are exempt from federal securities laws. See id. State insurance laws governing fixed annuity contracts require insurance companies to guarantee a minimum of the contract value after any costs and charges are applied. These 4 state laws generally require the minimum guarantee be at least 87.5 percent of the premiums paid, accumulated at an annual interest rate of 1 to 3 percent. Indexed Annuities and Certain Other Insurance Contracts (Final FIA Rule), 74 Fed. Reg. 3138, 3141 (Jan. 16, 2009) (to be codified at 17 C.F.R. Parts 230 and 240). The laws also generally impose disclosure and suitability requirements, which vary from state to state. A fixed index annuity (FIA) is a hybrid financial product that combines some of the benefits of fixed annuities with the added earning potential of a security. Like traditional fixed annuities, FIAs are subject to state insurance laws, under which insurance companies must guarantee the same 87.5 percent of purchase payments. Unlike traditional fixed annuities, however, the purchaser’s rate of return is not based upon a guaranteed interest rate. In FIAs the insurance company credits the purchaser with a return that is based on the performance of a securities index, such as the Dow Jones Industrial Average, Nasdaq 100 Index, or Standard & Poor’s 500 Index. Depending on the performance of the securities index to which a particular FIA is tied, the return on an FIA might be much higher or lower than the guaranteed rate of return offered by a traditional fixed annuity. Due to the fact that the purchaser’s actual return is linked to the performance of a securities index, however, the purchaser’s return cannot be calculated until the end of the crediting period. Insurance companies typically apply an annual crediting period; that is, the index-linked interest of an FIA is typically calculated on an annual basis after each one-year period ends. B. While this is the first case in which we have had occasion to address the § 3(a)(8) annuity exemption as it regards FIAs, the Supreme Court has offered guidance on the scope of the 5 exemption in VALIC, 359 U.S. 65, and SEC v. United Benefit Life Ins. Co., 387 U.S. 202 (1967). In VALIC, the Supreme Court considered whether a variable annuity fell within the § 3(a)(8) exemption. A variable annuity is a financial product under which purchasers pay premiums that are invested in common stocks and other equities to a greater degree than traditional annuities, and the benefit payments vary with the success of the investment management. See VALIC, 359 U.S. at 69. The Court explained that a variable annuity did not fall within the § 3(a)(8) exemption because it placed “all the investment risks on the [purchaser], none on the company.” Id. at 71. As the Court said, “the concept of ‘insurance’ involves some investment risk-taking on the part of the company.” Id. “‘[I]nsurance’ involves a guarantee that at least some fraction of the benefits will be payable in fixed amounts.” Id. Therefore, an issuer of an annuity “that has no element of a fixed return assumes no true risk in the insurance sense.” Id. The fact that there exists a risk of declining returns in difficult economic times is not sufficient to show that the insurer has assumed more risk under the contract. See id. Accordingly, because the variable annuity at issue did not offer a “true underwriting of risks, the one earmark of insurance,” the Court held that it did not fall within the exemption offered to traditional fixed annuities offered by insurers. Id. at 73. In a concurring opinion later approved by the full Court in United Benefit, Justice Brennan explained that when “a brand-new form of investment arrangement emerges which is labeled ‘insurance’ or ‘annuity’ by its promoters, the functional distinction that Congress set up in 1933 . . . must be examined to test whether the contract falls within the sort of investment form that Congress was then willing to leave exclusively to the State Insurance Commissioners.” Id. at 76 (Brennan, J., concurring); see United Benefit, 387 U.S. at 210. 6 In United Benefit, the Court concluded that another product similar to a variable annuity called a “Flexible Fund Annuity” was not exempt under § 3(a)(8) of the Act. A Flexible Fund functioned in much the same way as a variable annuity. Most notably, the purchaser paid premiums into a separate account that was primarily invested in common stocks, with the object of producing capital gains as well as an interest return. United Benefit, 387 U.S. at 205. Unlike the variable annuity in VALIC, however, the insurer guaranteed that the purchaser would receive a percentage of his premiums back. This percentage gradually increased from 50 percent of net premiums in the first year to 100 percent after 10 years. Id. United Benefit argued that, under VALIC, the existence vel non of substantial investment risk by the insurer ultimately determined whether a product fell within the § 3(a)(8) exemption. The Court disagreed that VALIC should be interpreted so narrowly. Id. at 210. Rather, the critical inquiry under § 3(a)(8) was whether the product at issue “‘involve[d] considerations of investment not present in the conventional contract of insurance.’” Id. (quoting Prudential Ins. Co. v. SEC, 326 F.2d 383, 388 (3d Cir. 1964)). In concluding that the Flexible Fund did not fall within § 3(a)(8), the Court relied significantly on the fact that the Flexible Fund “appeal[ed] to the purchaser not on the usual insurance basis of stability and security but on the prospect of ‘growth’ through sound investment management.” United Benefit, 387 U.S. at 211. Though the Court acknowledged that the “guarantee of cash value based on net premiums reduces substantially the investment risk of the contract holder,” it reasoned further that “the assumption of an investment risk [by the insurer] cannot by itself create an insurance provision under the federal definition.” Id. (citing Helvering v. Le Gierse, 312 U.S. 531, 542 (1941)). The Court recognized that a “basic difference” exists between “a contract which to some degree is insured and a contract of insurance.” 7 United Benefit, 387 U.S. at 211. In the case of the Flexible Fund, the insurer’s assumption of risk was minimal. The insurer was “obligated to produce no more than the guaranteed minimum at maturity, and this amount is substantially less than that guaranteed by the same premiums in a conventional deferred annuity contract.” Id. at 208. C. Since the Court’s decisions in VALIC and United Benefit, the SEC has engaged in rulemaking to address the newer financial products that have entered the market. In the mid- 1980s, the SEC promulgated Rule 151 in response to the creation of a new hybrid financial product called a guaranteed investment contract. See 17 C.F.R. § 230.151. Guaranteed investment contracts are like traditional fixed annuities, in that they promise a return at a guaranteed rate of return for the life of the contract. In some guaranteed investment contracts, however, the insurer may agree to periodically pay the purchaser an additional discretionary amount above the already guaranteed return amount. Rule 151 provided that, under certain conditions, a guaranteed investment contract would qualify for the § 3(a)(8) exemption notwithstanding an insurer’s ability to pay a discretionary amount to the purchaser. Under Rule 151, a contract falls within the § 3(a)(8) exemption if: (1) The annuity or optional annuity contract is issued by a corporation (the insurer) subject to the supervision of the insurance commissioner, bank commissioner, or any agency or officer performing like functions, of any State or Territory of the United States or the District of Columbia; (2) The insurer assumes the investment risk under the contract as prescribed in paragraph (b) of this section; 8 and (3) The contract is not marketed primarily as an investment. 17 C.F.R. § 230.151(a). Though the SEC considered excluding from the Rule 151 safe harbor any product in which an issuer calculates the rate of any excess return by reference to an index, it concluded that an issuer may reference an index to set the excess return rate, but only if the rate is set before each crediting period begins and remains in effect for at least one year. Definition of Annuity Contract or Optional Annuity Contract, S.E.C. Release No. 6645, 1986 WL 703849, at *11 (May 29, 1986). In the mid-1990s, insurance companies began marketing FIAs. The SEC did not take any regulatory action with respect to FIAs until 2007. By this time, the sales volume of FIAs had increased to $24.8 billion; indexed annuity assets totaled $123 billion. A total of 322 FIAs were being offered by 58 insurance companies. Having grown increasingly concerned that these FIAs were not being sold through registered broker-dealers and were not registered with the SEC despite their tie-in to a securities market, the SEC proposed Rule 151A. Rule 151A provides that a contract that is regulated as an annuity under state insurance law is not an “annuity contract” under § 3(a)(8) of the Act if: (1) The contract specifies that amounts payable by the issuer under the contract are calculated at or after the end of one or more specified crediting periods, in whole or in part, by reference to the performance during the crediting period or periods of a security, including a group or index of securities; and 9 (2) Amounts payable by the issuer under the contract are more likely than not to exceed the amounts guaranteed under the contract. 17 C.F.R. § 230.151A(a). By redefining an “annuity contract” to exclude FIAs, the Commission sought to ensure that purchasers of FIAs would be entitled to the full protection of the federal securities laws, including disclosure, antifraud, and sales practice protections. To support this new rule, the SEC first noted that the Securities Act did not define “annuity contract,” and that FIAs were not in existence at the time the “annuity contract” exemption in the Securities Act was enacted. Final FIA Rule, 74 Fed. Reg. at 3142-43. Without express statutory guidance, the Commission looked to the reasoning set forth in the Supreme Court’s decisions in VALIC and United Benefit to assess whether FIAs were the type of financial product that Congress would have been willing to leave to state insurance regulation. Id. at 3143. The SEC began its analysis by considering the level of risk associated with FIAs. Citing VALIC, the Commission reasoned that “Congress intended to include in the insurance exemption only those policies and contracts that include a ‘true underwriting of risks’ and ‘investment risk-taking’ by the insurer.” Id. (citing VALIC, 359 U.S. at 71-73). The annuities that were offered at the time of the enactment of the § 3(a)(8) exemption were fixed annuities that generally involved no investment risk to the purchaser. Final FIA Rule, 74 Fed. Reg. at 3143. Therefore, the SEC reasoned, Congress was willing to offer a securities law exemption to these types of no risk products because, by their nature, they did not raise the kinds of problems or risks that the federal securities laws were intended to address. Id. Additionally, the state insurance laws in 10 existence could adequately deal with any issues that might arise from such low risk insurance products. Id. On the other hand, the SEC explained, “[i]ndividuals who purchase [FIAs] are exposed to a significant investment risk–i.e., the volatility of the underlying securities index.” Id. at 3138. At the time an FIA is purchased, the purchaser “assumes the risk of an uncertain and fluctuating financial instrument, in exchange for participation in future securities-linked returns.” Id. at 3143. Unlike the guaranteed, fixed return offered by a traditional fixed annuity, the SEC asserted that an FIA’s return was neither known nor guaranteed. Id. The SEC acknowledged that “indexed annuities contracts provide some protection against the risk of loss,” but determined that these provisions did not adequately transfer the investment risk from the purchaser to the insurer. Id. Because the value of the purchaser’s investment was entirely dependent upon an unknown and fluctuating securities index, the assumption of a guaranteed minimum percentage of the FIA, though giving FIAs an outward aspect of insurance, was a superficial and unsubstantial offset of the purchaser’s risk. Id. Therefore, the SEC reasoned that an FIA’s value is much like that of a security, as the value of each product depends on the performance of the market. This securities-like investment risk, the SEC explained, was the exact type of investment risk that the Securities Act was created to address. Id. Though the SEC set forth in Rule 151 that the manner in which a product was marketed factored into the SEC’s determination of whether it constituted an annuity contract under § 3(a)(8), see 17 C.F.R. § 230.151(a)(3), the SEC opted not to include such an element in Rule 151A. The SEC noted that the performance of an FIA is obviously associated with the performance of a securities index, given the nature of the product. Final FIA Rule, 74 Fed. Reg. at 3146. Moreover, the SEC noted that the Supreme Court in VALIC did not consider how the variable annuity was marketed in determining whether 11 it fell within § 3(a)(8)’s exemption. Id. For these reasons, the SEC felt that inclusion of a marketing element in Rule 151A was unnecessary. Id. The SEC supplemented its analysis of Rule 151A by undertaking a consideration of the rule’s promotion of efficiency, competition, and capital formation, as is required by § 2(b) of the Act for certain SEC rulemakings. See 15 U.S.C. § 77b(b). The SEC first concluded that Rule 151A would promote efficiency, reasoning that the rule would extend the benefits of the disclosure and sales practice protections of the federal securities laws to FIAs that offered payments to the purchaser that fluctuated with the securities markets. Id. at 3169. The imposition of disclosure requirements would enable investors to make more informed investment decisions about purchasing FIAs, and would promote more suitable recommendations by issuers of FIAs to purchasers. Id. at 3169- 70. Next, the SEC asserted that the improvement in investors’ ability to make informed investment decisions would increase competition between issuers of FIAs. The SEC reasoned that the imposition of federal securities laws to regulate FIAs was particularly important because it would “bring about clarity in what has been an uncertain area of law,” id. at 3171, which would in turn increase competition because registered broker- dealers “who currently may be unwilling to sell unregistered [FIAs] because of their uncertain regulatory status may become willing to sell [FIAs] that are registered.” Id. at 3170. Finally, the SEC concluded that, based upon the increased efficiency resulting from the enhanced investor protections under federal law, Rule 151A would promote capital formation “by improving the flow of information among insurers that issue [FIAs], the distributors of those annuities, and investors.” Id. at 3171. Petitioners seek review of Rule 151A. 12 II. ANALYSIS A. Petitioners first argue that the SEC erred in excluding FIAs from the definition of “annuity contract” under § 3(a)(8) of the Act. Petitioners assert that their argument is supported by the plain language of the provision, as well as by the Supreme Court’s decisions in VALIC and United Benefit. Petitioners argue that Rule 151A is in conflict with the text of § 3(a)(8), the aforementioned decisions of the Court, and the text of the SEC’s prior rule, Rule 151. Finally, petitioners argue that the SEC failed to undertake properly its statutory responsibility to consider Rule 151A’s effect on efficiency, competition, and capital formation, pursuant to § 2(b) of the Act. We will address each argument in turn. When an agency is given express authority to execute and enforce its enabling statute and to prescribe such rules and regulations as are or may be necessary to carry out provisions of the statute, courts must apply a two-step analysis in reviewing the agency’s interpretation of the statute under Chevron U.S.A. Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837 (1984). Section 19(a) of the Act bestows upon the SEC the power to define terms and make rules to that effect. 15 U.S.C. § 77s(a). Rule 151A, which interprets the term “annuity contract” in § 3(a)(8) of the Act is clearly such a rule. Under Chevron, we first determine whether the statute being interpreted is ambiguous. If “Congress has directly spoken to the precise question at issue . . . [then] that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.” Chevron, 467 U.S. at 842-43. On the other hand, if the court determines that the statute is either “silent or ambiguous with 13 respect to the specific issue,” then Chevron Step One is satisfied. Id. at 843. Here, Chevron Step One is satisfied because the Act is ambiguous, or at the very least silent, on whether the term “annuity contract” encompasses all forms of contracts that may be described as annuities. Indeed, the analyses in the Supreme Court’s decisions in VALIC and United Benefit confirm this ambiguity. See generally VALIC, 359 U.S. 65; United Benefit, 387 U.S. 202. Had the statute been unambiguous, the Court need not have undertaken such an exhaustive inquiry in determining whether the two products at issue in those cases were annuities under § 3(a)(8) of the Act. Petitioners nevertheless argue that the Court’s decisions in VALIC and United Benefit establish that an “annuity” falls outside of § 3(a)(8) only if it is subject to the insurer’s investment management, and not subject to state insurance laws. Given the absence of these two elements, petitioners assert, § 3(a)(8) clearly governs FIAs because FIAs are not subject to the insurer’s investment management and are governed by a panoply of state insurance laws. Petitioners’ argument misses the mark because it interprets VALIC and United Benefit too restrictively. Nothing in those cases indicated that the Court’s determination whether the § 3(a)(8) exemption applies to particular contracts depends on the investment management of the issuer and the applicability of state insurance regulation. Rather, the Court embraced a broader approach in its § 3(a)(8) analysis. The Court clearly indicated in both VALIC and United Benefit that the § 3(a)(8) exemption applied to products that “‘did not present very squarely the sort of problems that the Securities Act . . . [was] devised to deal with, and which were, in many details, subject to a form of state regulation of a sort which made the federal regulation even less relevant.’” United Benefit, 387 U.S. at 210 (quoting VALIC, 359 U.S. at 75 14 (Brennan, J., concurring)). The Court therefore focused its § 3(a)(8) analysis on whether the product at issue “involve[d] considerations of investment not present in the conventional contract of insurance.” Id. (quotation omitted). Though an insurer’s investment management actions associated with a product may be relevant to determining whether that product is an annuity, this is not the only relevant characteristic. Petitioners’ reliance on the existence of state law regulation governing FIAs is also too limited. The Court recognized in United Benefit that it had “conclusively rejected” in VALIC the argument that the existence of adequate state regulation was the basis for the § 3(a)(8) exemption. Id. (quotation omitted). Therefore, the fact that FIAs are subject to state insurance regulation does not, without more, place them within the § 3(a)(8) exemption. Accordingly, the language of § 3(a)(8) does not unambiguously include FIAs within the § 3(a)(8) exemption. In light of the fact that the statute is ambiguous with respect to the term “annuity contract,” we reiterate that Chevron Step One is satisfied. We must next determine whether the SEC’s rule is a reasonable interpretation of the statute. The SEC’s rule will satisfy Step Two of the Chevron analysis so long as it meets this requirement. It is irrelevant that this court might have reached a different—or better—conclusion than the SEC. See Nat’l Cable & Telecomm. Ass’n v. Brand X Internet Servs., 545 U.S. 967, 980 (2005). In this case, the SEC has adopted an interpretation that is based in reason. By their nature, FIAs “appeal to the purchaser not on the usual insurance basis of stability and security but on the prospect of ‘growth’ through sound investment management.” United Benefit, 387 U.S. at 211. An FIA is akin to an annuity contract with respect to its pay-in and guaranteed minimum value of purchase payment features. The interest 15 return rate of an FIA, however, is decidedly more like a security in that the index-based return of an FIA is not known until the end of a crediting cycle, as the rate is based on the actual performance of a specified securities index during that period. Similar to an investor in securities, a purchaser of an FIA knows the level of annual return he will receive once the year is concluded and the index’s value is compared with its value at the beginning of the year. In FIAs, as in securities, there is a variability in the potential return that results in a risk to the purchaser. By contrast, an annuity contract falling under Rule 151’s exemption avoids this variability by guaranteeing the interest rate ahead of time. As these characteristics show, FIAs “involve considerations of investment not present in the conventional contract of insurance.” Id. at 210 (quotation omitted). Accordingly, the SEC’s interpretation that an FIA does not constitute an “annuity contract” under § 3(a)(8) of the Act was reasonable. Petitioners assert that the SEC based its analysis of Rule 151A on an “insupportable definition of investment risk.” The SEC determined that a purchaser bears sufficient risk to treat a product as a security when “[a]mounts payable by the issuer under the contract are more likely than not to exceed the amounts guaranteed under the contract.” 17 C.F.R. § 230.151A(a)(2) (emphasis added). In petitioners’ view, investment risk exists only where the purchaser of a security faces the possibility of a loss of principal. Petitioners’ view is certainly a defensible one. However, that is not sufficient to establish that the SEC’s rule is arbitrary or capricious. As the SEC points out, comparing two slightly different annuity products—one with a 5 percent interest rate guaranteed ahead of time; one with an interest rate that could be between 1 and 10 percent determined at the end of the year—the second product is riskier than the first product because its potential return could be lower than the rate of return from the first product, even 16 though it guarantees a minimum return rate of at least 1 percent. Moreover, the SEC has also been consistent in its position on investment risk. When it adopted Rule 151, which provided a safe harbor to certain types of annuity contracts including those that based interest payments on an investment index, the SEC noted that it was allowing products involving investment indices with the caveat that such index-based interest rates be calculated in advance of the upcoming year. See Definition of Annuity Contract or Optional Annuity Contract, S.E.C. Release No. 6645, 1986 WL 703849, at *11. This, the SEC said, would minimize the investment risk exposure of the purchaser. We cannot hold that this interpretation is unreasonable. Petitioners’ argument that the SEC failed to balance the investment risks assumed by the insurer against those assumed by the purchaser misses the mark. To the contrary, the SEC considered the risk to insurers, and weighed that risk against the risk to the FIA customer in determining whether the product is an annuity or security. See Final FIA Rule, 74 Fed. Reg. at 3143-50. The SEC concluded in its adopting release that annuities were those products that included a “true underwriting of risks” and “investment risk-taking” by the insurer, with more minimal risk exposure to the purchaser. Id. at 3143. In traditional fixed annuities, “the insurer bears the investment risk under the contract.” Id. at 3138. FIAs did not fall within this term because insurers offering FIAs left a more than minimal risk upon the purchaser. Rule 151A appears to be the SEC’s means of ensuring greater protection for consumers exposed to greater risk when insurers are exposed to less risk than normal. Indeed, the rule “sets forth a test” to distinguish between those contracts where the insurer bears more risk by paying a fixed amount, and those in which the purchaser bears more risk because he will receive a variable amount that is dependent upon fluctuating stock market prices. Final FIA Rule, 74 Fed. Reg. at 3145. Such an approach, in light of this risk assessment, is 17 reasonable. Petitioners’ argument that the SEC’s adoption of Rule 151A was in error because it allegedly failed to consider and balance the investment risks of the insurer and purchaser fails. Petitioners’ further argument that the SEC failed to account for marketing in considering whether a product is a security, though raising a closer issue, does not demonstrate that the SEC’s adoption of Rule 151A is unreasonable. Admittedly, the Supreme Court in United Benefit recognized that marketing is another significant factor in determining whether a state- regulated insurance contract is entitled to be exempt under § 3(a)(8). See United Benefit, 387 U.S. at 211 (contract found to “appeal to the purchaser not on the usual insurance basis of stability and security but on the prospect of ‘growth’ through sound investment management”). The SEC echoed this point when it enacted Rule 151 by stating that a product issued by a state-regulated insurance company falls within the § 3(a)(8) exemption if (1) the insurer assumes the investment risk of the contract and (2) “[t]he contract is not marketed primarily as an investment.” 17 C.F.R. § 230.151(a)(3) (emphasis added). Nevertheless, the SEC’s conclusion that an analysis of marketing was unnecessary for FIAs because “[i]t would be inconsistent with the character of such an indexed annuity, and potentially misleading, to market the annuity without placing significant emphasis on the securities-linked return and the related risks” is not unreasonable. Final FIA Rule, 74 Fed. Reg. at 3146. At the outset, the Supreme Court never held in either VALIC or United Benefit that marketing was an essential characteristic in assessing whether a product falls within the § 3(a)(8) exemption. Rather, as discussed above, the Court focused its inquiry on whether the product exhibited “considerations of investment not present in the conventional contract of 18 insurance.” United Benefit, 387 U.S. at 210 (quotation omitted); see also VALIC, 359 U.S. at 75 (Brennan, J., concurring) (§ 3(a)(8) exemption was to be considered a congressional declaration “that there then was a form of ‘investment’ known as insurance (including ‘annuity contracts’) which did not present very squarely the sort of problems that the Securities Act . . . [was] devised to deal with”). The fact that the Court considered how Flexible Funds were marketed only shows that that inquiry was relevant to Flexible Funds. It does not establish that the SEC must undertake a complete marketing analysis of FIAs in order to make a proper § 3(a)(8) determination. Regardless of that fact, the SEC did consider marketing of FIAs, and ultimately deemed the inclusion of this factor to be unnecessary. Again, we cannot say that the SEC’s decision was unreasonable. Indeed, the key characteristic of FIAs is that they offer a wide range of potential yearly return interest rates based on the performance of a securities index. This potential for a greater rate of return is what makes FIAs potentially more enticing than those exempt annuities that guarantee an interest rate ahead of time but at a lower rate. As we have noted above, this key distinction between these two products shows that FIAs are more like securities from a risk perspective than other annuity contracts. Where, as here, the essential characteristic of the product bestows upon that product obvious securities-like qualities, it is reasonable to assume that any marketing of the product would correspondingly be securities-related. It was also therefore reasonable for the SEC to find that the structure of FIAs–in particular the potential of securities-linked returns–is itself an implicit marketing tool aimed at consumers who wish to participate to some extent in the securities market, and that the rule need not contain an additional explicit prong addressed to marketing. Final FIA Rule, 74 Fed. Reg. at 3146. Indeed, this is in line with much of the analysis in United Benefit. See United Benefit, 387 U.S. at 211 (“‘The test . . . is what character the instrument is given in commerce by the terms of the offer, 19 the plan of distribution, and the economic inducements held out to the prospect.’”) (quoting SEC v. C.M. Joiner Leasing Corp., 320 U.S. 344, 352-53 (1943)). The SEC’s reasoning therefore was not in error. Contrary to petitioners’ arguments, Rule 151A does not conflict with Rule 151. Petitioners assert that Rule 151’s adopting release states that annuity contracts that have interest rates tied to a securities index can fall within the Rule 151 safe harbor so long as the rate of interest to be credited is not modified more frequently than once a year. They argue that FIAs fall within Rule 151’s safe harbor because the interest rate tied to the securities index is only determined once a year. This argument fails, however, because it ignores an express statement in the same portion of the adopting release of Rule 151 which states that annuity contracts that have interest rates tied to a securities index can fall within the Rule 151 safe harbor if the rate tied to the securities index is calculated prospectively. See Definition of Annuity Contract or Optional Annuity Contract, S.E.C. Release No. 6645, 1986 WL 703849, at *11 (“Once determined, the rate of excess interest credited to a particular purchase payment or to the value accumulated under the contract must remain in effect for at least the one-year time period established by the rule.” (emphasis added)). FIAs, however, do not calculate their rates of return tied to securities indices prospectively; their rates are calculated retroactively once the year is complete. Rule 151A is premised on this very distinction, and is therefore not in contradiction with Rule 151. For these reasons, we hold that the SEC’s interpretation of “annuity contract” was reasonable and Chevron Step Two is satisfied. 20 B. Even though the SEC’s interpretation of “annuity contract” was reasonable, petitioners argue that the SEC contravened § 2(b) of the Act because it failed to consider the efficiency, competition, and capital formation effects of the new Rule 151A. Section 2(b) of the Act states that, for every rulemaking in which the SEC “is required to consider or determine whether an action is necessary or appropriate in the public interest, the Commission shall also consider, in addition to the protection of investors, whether the action will promote efficiency, competition, and capital formation.” 15 U.S.C. § 77b(b). Petitioners argue that the costs of implementing Rule 151A are too burdensome and that the imposition of additional regulations would be inefficient. They also contend that the SEC failed to properly assess the existence of abuses of FIAs before applying securities regulations on the issuers of those products. The SEC counters that it properly rejected petitioners’ concerns regarding duplicative regulation because the Supreme Court’s decisions in VALIC and United Benefit made clear that state regulatory approaches to new products are not conclusive in a § 3(a)(8) analysis. In any event, the SEC argues that the challenges to its efficiency, competition, and capital formation analysis under § 2(b) fail because the SEC was not required to undertake such an analysis when it promulgated Rule 151A. At the outset, we must reject the SEC’s argument that no error occurred because the SEC was not required by the Securities Act to conduct a § 2(b) analysis. “The grounds upon which an administrative order must be judged are those upon which the record discloses that its action was based.” SEC v. Chenery Corp., 318 U.S. 80, 87 (1943). The SEC conducted a § 2(b) analysis when it issued the rule with no assertion that it was not required to do so. Therefore, the SEC must defend its analysis before the court upon the basis it employed in adopting 21 that analysis. See id. We now turn to the merits of the SEC’s § 2(b) analysis. We review the analysis under the statutory standard set by the Administrative Procedure Act. 5 U.S.C. § 706. The APA requires the court to set aside agency action that is “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” Id. at § 706(2)(A). We hold that the Commission’s consideration of the effect of Rule 151A on efficiency, competition, and capital formation was arbitrary and capricious. The SEC purports to have analyzed the effect of the rule on competition, but does not disclose a reasoned basis for its conclusion that Rule 151A would increase competition. The SEC concluded that enacting the rule would resolve the present uncertainty prevailing over the legal status of FIAs. The SEC reasoned that the rule “will bring about clarity in what has been an uncertain area of law.” Final FIA Rule, 74 Fed. Reg. at 3171. The SEC explained that this newfound “clarity” brought about by the rule would enhance competition because insurers who may have been reluctant to issue indexed annuities, while their status was uncertain, may decide to enter the market. Similarly, registered broker-dealers who currently may be unwilling to sell unregistered indexed annuities because of their uncertain regulatory status may become willing to sell indexed annuities that are registered, thereby increasing competition among distributors of indexed annuities. Id. at 3170. This reasoning is flawed. The lack of clarity resulting from the “uncertain legal status” of the financial product is only another way of saying that there was not a regulation in place 22 prior to the adoption of Rule 151A determining the status of those products under the annuity exemption of § 3(a)(8). The SEC cannot justify the adoption of a particular rule based solely on the assertion that the existence of a rule provides greater clarity to an area that remained unclear in the absence of any rule. Whatever rule the SEC chose to adopt could equally be said to make the previously unregulated market clearer than it would be without that adoption. Moreover, the fact that federal regulation of FIAs would bring “clarity” to this area of the law is not helpful in assessing the effect Rule 151A has on competition. Again, creating a rule that resolves the “uncertain legal status” of FIAs might be said to improve competition. But that conclusion could be asserted regardless of whether the rule deems FIAs to fall within the SEC’s regulatory reach or outside of it. Indeed, the SEC would achieve a similar clarity if it declined outright to regulate FIAs. Section 2(b) does not ask for an analysis of whether any rule would have an effect on competition. Rather, it asks for an analysis of whether the specific rule will promote efficiency, competition, and capital formation. 15 U.S.C. § 77b(b). The SEC’s reasoning with respect to competition supports at most the conclusion that any SEC action in this area could promote competition, but does not establish Rule 151A’s effect on competition. Section 2(b) requires more than this. See id. The SEC’s competition analysis also fails because the SEC did not make any finding on the existing level of competition in the marketplace under the state law regime. The SEC asserted competition would increase based upon its expectation that Rule 151A would require fuller public disclosure of the terms of FIAs and thereby increase price transparency. The SEC could not accurately assess any potential increase or decrease in competition, however, because it did not assess the baseline level of price transparency and information disclosure under state law. The SEC nevertheless argues that it is not required to 23 conduct such a detailed § 2(b) analysis because doing so would contravene the Supreme Court’s reasoning in United Benefit and VALIC. According to the Commission, these two cases established that adequate state regulation is not relevant to whether a product qualifies for a § 3(a)(8) exemption. See United Benefit, 387 U.S. at 210 (“The argument that the existence of adequate state regulation was the basis for the exemption [under § 3(a)(8)] was conclusively rejected in VALIC.” (quotation omitted)); VALIC, 359 U.S. at 75 (Brennan, J., concurring) (“[H]owever adequately State Securities Commissioners might regulate an investment, it was not for that reason to be freed from federal regulation. Concurrent regulation, then, was contemplated by the Acts as a quite generally prevailing matter.”). Therefore, the SEC argues, it was reasonable to conclude that state regulation, “no matter how strong,” could not “substitute for the federal securities law protections that apply to instruments that are regulated as securities.” Final FIA Rule, 74 Fed. Reg. at 3170. The SEC’s reliance on VALIC and United Benefit is misplaced. The SEC’s obligations under § 2(b) are distinct from the questions posed in VALIC and United Benefit. Those cases addressed whether a particular product fell within the § 3(a)(8) exemption; § 2(b) imposes on the SEC an obligation to consider the economic implications of certain rules it proposes. See Chamber of Commerce v. SEC, 412 F.3d 133, 143 (D.C. Cir. 2005). Accordingly, the SEC’s § 2(b) analysis is arbitrary and capricious because it failed to consider the extent of the existing competition in its analysis. The Commission’s efficiency analysis is similarly arbitrary and capricious. The SEC concluded that Rule 151A would promote efficiency because the required disclosures under the rule would enable investors to make more informed investment decisions about purchasing indexed annuities. The SEC advanced further that the rule’s sales practice protections would 24 enable sellers to promote more suitable recommendations to investors; this, in turn, would lead to investors making even better informed decisions, which would offer greater efficiency. As with its analysis of competition, however, the SEC’s analysis is incomplete because it fails to determine whether, under the existing regime, sufficient protections existed to enable investors to make informed investment decisions and sellers to make suitable recommendations to investors. The SEC’s failure to analyze the efficiency of the existing state law regime renders arbitrary and capricious the SEC’s judgment that applying federal securities law would increase efficiency. Finally, the SEC’s flawed efficiency analysis also renders its capital formation analysis arbitrary and capricious. The SEC’s conclusion that Rule 151A would promote capital formation was based significantly on the flawed presumption that the enhanced investor protections under Rule 151A would increase market efficiency. This analysis fails with the failure of its underlying premise. Having determined that the SEC’s § 2(b) analysis is lacking, we conclude that this matter should be remanded to the SEC to address the deficiencies with its § 2(b) analysis. See Chamber of Commerce, 412 F.3d at 145 (citing Fox Television Stations, Inc. v. FCC, 280 F.3d 1027, 1048-49 (D.C. Cir. 2002)). It is obvious that the SEC believes imposing a federal framework on FIAs would be superior to the existing patchwork of state insurance laws. Indeed, after a more thorough review of the existing state law regime, the Commission may decide ultimately that Rule 151A will promote competition, efficiency, and capital formation. Nevertheless, the Commission must either complete an analysis sufficient to satisfy its obligations under § 2(b), or explain why that section does not govern this rulemaking. Accordingly, we grant the petitions and remand for further reconsideration consistent with this opinion. So ordered.