Securities regulation in the shadow of the antitrust laws: the case for a broad implied immunity doctrine.
NOTE CONTENTS INTRODUCTION I. A DOCTRINAL DEFENSE OF BILLING A. The Trio of Pre-Billing Implied Immunity Cases B. The Decision in Billing C. Billing Is Consistent with Precedent II. THE RISK OF OVERDETERRENCE A. The False Positives Concern 1. The SEC's Comparative Advantage 2. Counterarguments to the SEC's Comparative Advantage a. Is the SEC Focused on Competition? b. Agency Capture c. The Unavailability of Treble Damages B. The Excessive Liability Concern 1. The Competitive Nature of the Securities Markets 2. The Effect of the Variance of Antitrust Damage Awards III. THE OPTIMAL IMPLIED IMMUNITY STANDARD A. The Effect of Time Constraints on SEC Regulation B. The Ability of Courts To Interpret SEC Regulations Correctly CONCLUSION
In Credit Suisse Securities (USA) LLC v. Billing, (1) the Supreme Court held that an antitrust suit challenging various alleged concerted marketing activities of underwriters of initial public offerings (IPOs) was impliedly precluded by the securities laws. The Court found the underwriters, whose actions included requiring customers to agree to purchase additional securities following the IPO, immune from antitrust liability based on the incompatibility of the securities laws and antitrust laws, even though the Securities and Exchange Commission (SEC) had condenmed much of the conduct alleged in the plaintiffs' antitrust complaint. (2) The Court reasoned that the distinction between activities that the SEC permits and those that it prohibits can be very fine and is subject to change and that "nonexpert" judges and juries are likely to have a comparative disadvantage in determining on which side of the line a particular activity falls. (3) In the Court's view, the possibility that courts adjudicating antitrust claims ("antitrust courts") might "make unusually serious mistakes" in this area would have a chilling effect on the securities industry, (4) and this justified giving the SEC exclusive jurisdiction over the underwriting activities at issue. (5)
The decision in Billing has generated significant criticism. A number of commentators have argued that the Court was wrong to focus on the potential for erroneous decisions in antitrust cases even when the SEC has not explicitly given the conduct in question its imprimatur or, as in Billing itself, when the SEC has in fact prohibited the conduct. (6) These critics have argued that a finding of implied immunity cannot be predicated merely on the SEC's jurisdiction over, or even review of, a particular class of conduct unless the SEC affirmatively permits the conduct and thereby immunizes it from an antitrust challenge. (7) Several commentators have also expressed concern that, as a policy matter, the preclusion of antitrust suits may lead to the underdeterrence of anticompetitive conduct in the securities industry. (8)
This Note responds to such criticism by offering both a doctrinal and a normative defense of the Court's implied immunity analysis in Billing. The Note proceeds in three Parts. Part I presents a doctrinal argument in support of Billing. It contends that critics of the decision have mischaracterized the relevant precedents and have invoked untenable bases on which to distinguish them from Billing. Instead, it argues that the Court's interpretation of the securities laws as impliedly precluding antitrust suits even in the absence of a manifest conflict between substantive securities and antitrust law is consistent with Supreme Court precedent in this area. (9)
The subsequent two Parts map out a normative argument in support of Billing's broad implied immunity standard. Part II argues that the SEC, which has an obligation to consider competition effects when promulgating regulations, (10) possesses a comparative advantage over antitrust courts in determining the scope of permissible conduct in the securities industry and that the latter can be expected both to prohibit socially beneficial conduct and to impose excessive liability, even for activities that should be prohibited. The risk of false positives in antitrust cases stems from antitrust courts' relative lack of securities expertise and antitrust law's narrow focus on competition to the exclusion of other legitimate policy goals in the securities area. (11) In addition, the competitive nature of the securities industry (12) and the greater variability of antitrust jury awards as compared to analogous penalties in SEC enforcement actions (13) suggest that antitrust courts may impose excessive damages even in cases in which they correctly determine that an activity merits prohibition.
Part III draws on the inferences from Part II to argue for the efficiency of a broad implied immunity doctrine under which antitrust suits are precluded whenever the SEC has jurisdiction over a particular activity and is actively engaged in reviewing its merits. But whereas the Court's opinion in Billing focused primarily on the possibility that antitrust courts might reach the wrong result ex post, Part III shifts the focus to the SEC's regulatory decisions ex ante. In particular, it argues that under a narrower implied immunity standard, the SEC might forgo a superior and more nuanced regulatory approach in favor of a blanket authorization of a particular practice in order to preempt errors by antitrust courts. The principal benefit of the Court's broad grant of immunity in Billing is that it frees the SEC from having to regulate in the shadow of the antitrust laws in this manner. Paradoxically, this analysis also suggests that a broad implied immunity standard may actually lead to more antitrust enforcement than would a narrower rule.
I. A DOCTRINAL DEFENSE OF BILLING
Billing has spawned a fair bit of academic criticism. The principal critique leveled against the decision is that the Court was incorrect in finding the securities laws and antitrust laws incompatible given that both regimes appeared to condemn the challenged conduct. (14) This Part argues that such criticism is based on a mischaracterization of the relevant Supreme Court precedents, which instruct that SEC disapproval of an activity is not dispositive to the implied immunity analysis. Billing is therefore consistent with prior cases.
A. The Trio of Pre-Billing Implied Immunity Cases
Prior to Billing, the Supreme Court had decided three cases involving assertions of implied antitrust immunity under the securities laws. The first was Silver v. New York Stock Exchange. (15) Silver involved a decision by the New York Stock Exchange to prohibit the use of direct telephone wire connections between exchange members and nonmember broker-dealers. A nonmember brokerage that was unable to obtain price quotations quickly as a result of the rule alleged that the prohibition was a conspiracy in restraint of trade in violation of the Sherman Act. (16) The Court first explained that the removal of the wires would normally constitute a per se violation of section 1 of the Act, since it functioned as a group boycott. (17) However, the presence of a parallel regulatory scheme in the Securities Exchange Act of 1934, and its policy of self-regulation by the national exchanges, meant that the antitrust laws could be applied only if they were reconcilable with the securities laws. (18) Emphasizing the "'cardinal principle of construction that repeals by implication are not favored,'" the Court explained that "[r]epeal is to be regarded as implied only if necessary to make the Securities Exchange Act work, and even then only to the minimum extent necessary." (19) At the time, the Exchange Act required exchanges to register with the SEC and to submit a copy of their rules, which were required to be "'just and adequate to insure fair dealing and to protect investors.'" (20) Although the SEC had the power to disapprove of an exchange's rules, it did not have the authority to review particular instances of the Exchange's enforcement of those rules. (21) Because the SEC lacked such jurisdiction, it was incapable of performing an "antitrust function" sufficient to displace the antitrust laws. (22) Thus, the Court declined to read an implied repeal of the antitrust laws into the Exchange Act. But the Court emphasized the limited reach of its decision, commenting that "[w]ere there Commission jurisdiction and ensuing judicial review for scrutiny of a particular exchange ruling ... a different case would arise." (23)
Just over a decade later, the Court was presented with such a case. Gordon v. New York Stock Exchange involved a challenge under sections 1 and 2 of the Sherman Act to the fixing of the brokerage commission rates charged by members of the New York Stock Exchange for smaller transactions. (24) As recently amended, the Exchange Act contained a general prohibition against the fixing of commission rates by an exchange but also empowered the SEC to make exceptions permitting an exchange to fix commissions provided that the rates set were reasonable in relation to brokers' costs and did not impose an unnecessary burden on competition. (25) The SEC, moreover, had been continuously engaged in the process of reviewing the practice of rate fixing by exchanges. (26) The Court held that the antitrust claims were impliedly precluded by the securities laws and declined to issue an injunction prohibiting the Exchange from fixing commissions going forward. It distinguished Silver on the ground that the Exchange Act gave the SEC explicit regulatory power to review exchange rules fixing brokers' commission rates, and the SEC had engaged in such review during the preceding years. (27) Given the SEC's clear jurisdiction to regulate the conduct at issue, the Court expressed concern that if the antitrust suit were permitted to proceed, then the exchanges and their members might be subject to conflicting standards. The likely cause of a conflict, the Court reasoned, was that, while the antitrust laws' exclusive objective is to promote competition, the securities laws have multiple purposes, including "the economic health of the investors, the exchanges, and the securities industry." (28) Thus, even though the SEC's position at the time was that fixed commission rates should be abolished, (29) the possibility of a conflict in the future was sufficient to imply a repeal of the antitrust laws.
The third Supreme Court decision addressing implied antitrust immunity in the securities context is United States v. National Ass'n of Securities Dealers, Inc. (30) (NASD), decided the same day as Gordon. In NASD, the government and investors brought suit against various mutual funds and dealers under section 1 of the Sherman Act. The complaint alleged that the mutual funds, in an attempt to inhibit the development of a secondary market for the funds' securities, had engaged in resale price maintenance by fixing the prices at which broker-dealers could purchase or sell a fund's shares from or to investors. The complaint also alleged that the mutual funds had prohibited broker-dealers from selling shares to other dealers. (31)
The Court again found that the securities laws impliedly immunized the mutual funds' restrictions from antitrust liability. The Investment Company Act authorized mutual funds to impose restrictions on sales of their shares provided that such restrictions were consistent with their registration statements and with regulations that the SEC was authorized to promulgate in the interests of the funds' shareholders. (32) However, the SEC had not exercised its rulemaking power to prescribe any such standards. (33) Nevertheless, the Court concluded that the statute reflected a determination by Congress that "subject to Commission oversight, mutual funds should be allowed to retain the initiative in dealing with the potentially adverse effects of disruptive trading practices." (34) In the Court's view, the SEC's decision to accept the restrictions imposed by mutual funds on the transferability of their shares did not constitute abdication of its oversight role but rather reflected an informed judgment that these restrictions were appropriate. (35) Thus, although at that time the vertical price restraints would constitute per se violations of section 1 of the Sherman Act, (36) the Court concluded that the antitrust laws were irreconcilable with the regulatory scheme established by the Investment Company Act (37) and that implied immunity was necessary to enable the SEC to do its job "free from the disruption of conflicting judgments that might be voiced by courts exercising jurisdiction under the antitrust laws." (38) Importantly, the Court reached this conclusion despite the fact that the SEC had recently asked the NASD to amend its rules to prohibit resale price maintenance agreements between mutual fund underwriters and broker-dealers. (39) Thus, it implied immunity even though, at the time, the SEC apparently disapproved of the challenged conduct.
B. The Decision in Billing
Billing was the first implied immunity case implicating the securities laws that the Supreme Court decided in the more than thirty years after Gordon and NASD. It involved an antitrust suit challenging various alleged agreements among a number of investment banks regarding the underwriting of IPOs. (40) The plaintiffs contended that the underwriters required investors to (1) purchase shares in the aftermarket following the IPO, a practice known as "laddering"; (2) commit to purchase other "less attractive" securities from the underwriters, an arrangement generally referred to as "tying" in antitrust parlance; and (3) agree to purchase the issuer's shares in subsequent public offerings, which would generate additional commissions for the underwriters. (41)
In assessing the defendants' implied immunity argument, the Court distilled from its precedents four factors relevant to the determination of whether the antitrust laws and the securities laws are "clearly incompatible" in a particular context: (1) whether the securities laws give the SEC authority to supervise the conduct at issue; (2) whether the SEC in fact exercises that authority; (3) the resulting risk that the antitrust laws and the securities laws might produce conflicting "guidance, requirements, duties, privileges, or standards of conduct"; and (4) whether the potential conflict affects practices that lie in the heartland of financial market activity that the securities laws seek to regulate. (42)
Applying these factors to the activities at issue in the case, the Court concluded that factors one, two, and four were clearly satisfied. First, the SEC has jurisdiction over the sales practices of underwriters by virtue of its power to regulate communications between underwriters and their customers and to prohibit fraudulent, deceptive, or manipulative practices. (43) The SEC, moreover, had exercised its authority to regulate IPO sales by promulgating regulations defining the permissible scope of underwriter sales efforts during their marketing campaigns. As to the fourth factor, the Court emphasized that the IPO process "is central to the proper functioning of well-regulated capital markets" and "lie[s] at the very heart of the securities marketing enterprise." (44) Thus, the outcome of the case turned on the third factor-whether an antitrust suit would be incompatible with the SEC's administration of the securities laws.
The Court found that the securities laws and the antitrust laws were indeed incompatible in this context, despite the fact that the SEC had issued guidance disapproving of much of the conduct alleged in the complaint. The Court's incompatibility analysis invoked two distinct concerns. First, the Court suggested that even if the SEC at present disapproved of the conduct at issue, there was no guarantee that the SEC would not change its position in the future. (45) Second, even assuming that the SEC would continue to disapprove of such conduct, the Court reasoned that permitting an antitrust suit still might lead to inconsistent results for a number of reasons. (46)
In the Court's view, the most likely source of this inconsistency is the fine line between marketing activities that the SEC considers permissible and activities that it forbids. For example, underwriters are prohibited from soliciting aftermarket orders prior to the completion of the IPO, but they are permitted to ask customers about their longer-term plans to purchase additional shares. (47) Similarly, although the SEC prohibits explicit tying arrangements in which an underwriter demands that its customers purchase additional securities, it permits firms to allocate IPO shares to customers that have previously purchased other services from the firm at a reasonable rate. (48) The Court reasoned that antitrust courts are likely to have difficulty discerning the precise contours of the boundary that separates permissible from impermissible conduct in this area.
The possibility of erroneous judicial decisions was heightened by two additional factors. First, the same piece of evidence might be interpreted to support a finding of impermissible behavior or permissible marketing activity under the SEC's standards. The Court gave as an example a conversation in which an underwriter inquires as to the investor's anticipated holding period; such a question might be evidence of impermissible laddering or of merely a benign attempt to stabilize the aftermarket share price by allocating IPO shares to investors who plan on holding them for a longer period of time. (49) Second, in light of the fine line separating permissible from impermissible conduct in this area, different courts are likely to reach inconsistent decisions in like cases. In combination, these factors suggested that antitrust courts can be expected to make significant mistakes, which would chill regulated entities from engaging in socially beneficial activities and impair the functioning of the securities markets. (50)
Finally, the Court noted that the benefits from antitrust enforcement tend to be small when the SEC is actively engaged in regulating a particular activity. The Court emphasized that the SEC is required to consider competition effects when it exercises its rulemaking authority. (51) Because the SEC, aided in its enforcement efforts by private securities lawsuits, can adequately police anticompetitive behavior among underwriters, the Court concluded that the marginal benefit from antitrust suits is minimal and significantly outweighed by the possibility of judicial error.
C. Billing Is Consistent with Precedent
Billing's implied immunity analysis generally comports with the precedents discussed above. As an initial matter, the four-factor test articulated in Billing places significant emphasis on the scope of the SEC's regulatory authority to approve or proscribe the conduct at issue. This is the principal factor that distinguishes Silver, in which the Court declined to imply immunity because the SEC lacked legal authority to veto specific exchange rules, from Gordon and NASD, in which the Court found immunity where the SEC had the power to approve or prohibit the activities at issue. Of Billing's four factors, factors one and two entail a direct inquiry into the scope of the SEC's regulatory authority and whether the SEC has exercised that authority. Factor four, which looks to whether the conduct at issue falls into the heartland of securities activity, is also a kind of proxy for the extent of SEC oversight.
In addition, Billing's holding that implied immunity does not require the SEC to have affirmatively approved a particular activity is consistent with both NASD and Gordon. In NASD, the Court found the antitrust and securities laws irreconcilable even though the SEC had not promulgated standards against which to evaluate the restrictions imposed by mutual funds with regard to the resale of their shares and had recently expressed disapproval of the vertical restraints at issue in the case. (52) And Gordon implicitly recognized that even though the SEC, at the time, condemned rate-fixing among exchange members, the possibilities that the SEC might regulate in this area in the future and that courts might reach different results from those reached by the SEC created clear discord between the securities laws and the antitrust laws. (53)
Moreover, both Billing and Gordon expressed a similar concern that antitrust courts would prohibit too much conduct and thereby deter socially beneficial activities in the securities industry. The two decisions, however, offered slightly different explanations for the source of such false positives. In Gordon, the Court suggested that the problem was one of differing priorities: antitrust courts would consider only competition, whereas the SEC's analysis is multidimensional and incorporates the well-being of investors, firms, and the exchanges as well. By contrast, Billing's implied immunity analysis focused on judicial competence to evaluate the relevant evidence and apply SEC regulations correctly. But in both cases the Court's finding of a conflict ultimately rested on a common premise--that, compared to the SEC, courts are poorly equipped to determine the scope of permissible conduct in the securities industry.
Nevertheless, Billing has been criticized for unduly broadening the implied immunity doctrine by failing to reconcile the securities laws and the antitrust laws despite the absence of a clear substantive conflict with respect to the two regimes' treatment of the conduct at issue in the case. (54) These criticisms founder on several points.
To begin with, Billing's critics rely to a great extent on the canon of construction, acknowledged in Silver and in each of the three cases thereafter, that implied repeals of the antitrust laws are disfavored and should be found "'only if necessary to make the Securities Exchange Act work, and even then only to the minimum extent necessary.'" (55) While rhetorically powerful, this canon is not a formula for deciding concrete cases and mere recitation of the canon does not illuminate the factors that the Court has used to determine whether a particular securities activity is impliedly immune from antitrust liability. Critics of Billing have attempted to demonstrate the supposedly insuperable effect of the canon by pointing out the infrequency with which the Court has interpreted statutes to impliedly immunize an activity in a regulated industry from antitrust suit. (56) This mode of analysis is problematic for several reasons.
First, although the Court has often declined to read regulatory statutes impliedly to repeal the antitrust laws, it has also found immunity outside of the securities context even when the challenged conduct was unlawful under the applicable regulatory statute. In Pan American World Airways v. United States, the Court held that an antitrust complaint alleging a conspiracy among multiple air carriers to divide territories and limit routes should be dismissed even though the Civil Aeronautics Board disapproved of the defendants' conduct and had in fact requested that the Attorney General bring suit. (57) The Court reasoned that whether a transaction meets the "standards of competition and monopoly" set forth in the Federal Aviation Act is "peculiarly a question for the Board" (58) and expressed concern that "[i]f the courts were to intrude independently with their construction of the antitrust laws, two regimes might collide." (59) Pan American thus confirms the principle illustrated by both Gordon and NASD that regulatory approval of a challenged activity is not a prerequisite for finding implied antitrust immunity.
More importantly, the inference that critics of Billing draw from the fact that findings of implied immunity are relatively rare is misleading insofar as it ignores the industry-specific nature of the Supreme Court's implied immunity jurisprudence. The issue as to whether the securities laws effect an implied repeal of the antitrust laws with respect to certain activities is a question of statutory interpretation, one that focuses narrowly on the meaning of the securities laws; the Court's interpretation of other regulatory statutes is largely inapposite. As then-Judge Anthony Kennedy explained:
[T]here is no simplistic and mechanically universal doctrine of implied antitrust immunity; each of the Supreme Court's cases is decisively shaped by considerations of the special aspects of the regulated industry involved.... [T]he uncritical transfer of abstract characterizations about the implied immunity of one industry to the different circumstances of another industry is not a reliable method of analysis. (60)
Thus, the precedential value of cases in which the Court has declined to imply immunity outside of the securities context is minimal. In each of those cases, the Court concluded, often on the basis of legislative history, that Congress had not intended to create a pervasive regulatory scheme to displace the antitrust laws. (61) As the preceding Sections make clear, the Court has generally reached the opposite conclusion in interpreting the securities laws.
Indeed, in illustrating the differences in the Court's implied immunity cases across various industries, then-Judge Kennedy singled out the securities industry as one that has enjoyed broad antitrust immunity. The Court's "reluctan[ce] to allow via the antitrust laws any tampering with the regulatory framework" established by the securities laws was justified, in Kennedy's view, by the particular circumstances surrounding their adoption, "the grave historical crises caused by the absence of regulation in those industries." (62) In addition, he noted that the Court's implied immunity jurisprudence in the securities area was influenced by the securities laws' reliance on self-regulation, which would be frustrated by "subjecting to antitrust liability the rulemaking and enforcement functions Congress charged the industry with performing." (63) Thus, notwithstanding the presumption against implied repeals and the Court's frequent refusal to interpret other regulatory statutes to preempt antitrust litigation, the Court has been decidedly receptive to claims of implied immunity under the securities laws.
Commentators' efforts to discredit Billing on the basis of relevant securities precedents are also unavailing. In particular, some have attempted to distinguish Billing from NASD and Gordon on the ground that, in Billing, the SEC disapproved of the activities at issue and thus there was no conflict between the securities laws and the antitrust laws. (64) This argument is based on an incorrect reading of both Gordon and NASD as cases in which the SEC approved of the activities being challenged in the antitrust action. (65) Although in both cases the SEC had previously permitted the conduct at issue, at the time the cases were decided, it disapproved of the challenged conduct. (66) Nevertheless, in each case, the Court rejected the plaintiffs' request for an injunction. (67)
Critics of Billing have also tried to distinguish the case on the grounds that the Exchange Act does not expressly give the SEC the specific authority to authorize the alleged tying and laddering agreements at issue; (68) rather, the Commission's regulatory authority stems from its general powers to prohibit fraudulent and manipulative practices and to regulate underwriters' solicitation of and communications with customers. (69) But there are two problems with this argument. First, although the specificity of the statutory provision at issue might be a basis for distinguishing Billing from Gordon, because in the latter case the statute explicitly prohibited the exchanges from imposing fixed commissions subject to the power of the SEC to make exceptions, (70) it is not clearly a basis for distinguishing Billing from NASD. The provision of the Investment Company Act at issue in NASD prohibited mutual funds from "'restrict[ing] the transferability or negotiability of any security of which it is the issuer'" unless the restrictions were disclosed in its registration statement and were consistent with regulations that "'the Commission may prescribe in the interests of the holders of all of the outstanding securities of such investment company.'" (71) This is a fairly broad delegation of rulemaking power to the SEC and is not so precisely worded as to constitute an explicit authorization to the SEC to permit the resale price maintenance restrictions at issue in the case.
The second and more fundamental problem with this argument is that it does not represent a principled approach to distinguishing between statutory provisions of the securities laws for implied immunity purposes. The securities laws contain many broadly worded provisions delegating expansive rulemaking authority to the SEC. (72) These delegations evince a congressional intent to give the SEC significant flexibility in promulgating regulations in the public interest. To fail to give effect to such provisions for the purposes of determining whether the securities laws and antitrust laws are incompatible with respect to a particular activity would therefore frustrate, not advance, the intent of Congress. Thus, the degree of specificity of the statutory mandate at issue in Billing is not a compelling basis for distinguishing it from Gordon or NASD.
Ultimately, the argument that Billing represents a significant break from precedent is not persuasive. To be sure, Billing's manifest concern with the ability of antitrust courts to apply SEC regulations correctly constitutes a new explanation as to why present harmony between substantive securities regulations and antitrust law does not preclude a finding of implied immunity. But introducing a new justification for a preexisting doctrine hardly violates principles of stare decisis. Indeed, the reasoning in Billing is consistent with the Court's overall antitrust jurisprudence over the last several decades, which has reflected an increased concern about false positives in antitrust cases. (73)
II. THE RISK OF OVERDETERRENCE
Although the principal criticisms of Billing have been doctrinal, several commentators have also expressed concern that, as a result of the immunization of certain securities activities from antitrust suits, anticompetitive conduct in the securities industry will frequently go unpunished and underdeterred. (74) This Part argues that such a concern is unfounded, that the SEC is better positioned than antitrust courts to distinguish socially beneficial securities activities from inefficient anticompetitive conduct, and that permitting parallel antitrust suits would in fact lead to overdeterrence, not underdeterrence.
There are two distinct components to this argument. The first Section argues that antitrust courts are likely to prohibit too much conduct in the securities area because, unlike the SEC, they lack securities-related expertise and may be unable to balance competition concerns against other considerations that are relevant to the vitality of the securities markets. The second Section argues that because of the competitive nature of the securities industry and the greater variance of jury awards in antitrust cases as compared to SEC sanctions, antitrust courts on average can be expected to impose excessive liability even as to activities that both the securities laws and the antitrust laws condemn.
A. The False Positives Concern
1. The SEC's Comparative Advantage
There are several reasons to prefer that the scope of permissible conduct in the securities industry be determined by the SEC rather than by antitrust courts. First, SEC regulation serves a sufficient antitrust function. The Commission is required to take into consideration competition effects when it promulgates regulations (75) and, when promulgating a regulation pursuant to its authority under the Exchange Act, it must explain the effect of the regulation on competition in its statement of basis and purpose. (76) Because antitrust principles permeate the securities laws, the utility of parallel antitrust enforcement is dubious.
Moreover, SEC regulation possesses two affirmative advantages over antitrust litigation. First, the SEC is a specialized agency with a great deal of expertise in the securities area, whereas antitrust courts are generalized tribunals and antitrust matters occupy only a small portion of their dockets. (77) In addition, when promulgating and enforcing its regulations, the SEC can take an industry-wide perspective and can alter its position on the basis of new information. By contrast, judicial decisions are limited to the facts of the particular case at bar and are comparatively less dynamic.
Second, relative to SEC regulation, antitrust analysis is comparatively narrow in scope. Whereas antitrust courts focus exclusively on the effect of an activity on competition, (78) SEC regulation takes into account, in addition to competition, the effect of a potential rule on the volatility of the capital markets, the accuracy of securities pricing, fraudulent practices by broker-dealers, and the health of regulated companies. (79) To the extent that these goals might, at times, conflict with the paradigm of unfettered competition," (80) the SEC is in a better position to strike the right balance among them than are antitrust courts. (81) In particular, because antitrust courts can be expected to undervalue the non-competition-related benefits of a given activity, they are likely to prohibit some conduct that should be permitted. For example, price stability, a policy which is germane to securities regulation, (82) is potentially in tension with traditional antitrust principles. (83) As such, antitrust courts may impose liability for concerted action designed to stabilize securities prices even if such action is on balance beneficial. (84)
Indeed, the SEC's comparative advantage over antitrust courts is implicit in the uncontroversial application of the implied antitrust immunity doctrine to practices that the Commission has expressly approved. (85) In such cases, the antitrust laws must give way to SEC regulation because Congress has, by giving the SEC the authority to reach a result that might be inconsistent with the antitrust laws, expressed its preference for SEC regulation over antitrust actions. (86) Nevertheless, the following Subsection considers several arguments that the SEC may not in fact be better equipped than antitrust courts to determine the scope of permissible conduct in the securities industry.
2. Counterarguments to the SEC's Comparative Advantage
Commentators have presented various counterarguments challenging the assumption that the expertise of the SEC and its ability to pursue multiple regulatory goals make it a more competent antitrust regulator than the courts. This Subsection will examine three such arguments, in particular: (2) that the SEC does not devote sufficient resources to policing competition; (2) that the SEC may be overly lax in its antitrust regulation because of agency capture; and (3) that the inability of the SEC, unlike antitrust courts, to impose treble damages makes it a comparatively impotent regulator. None of these arguments is sufficient to rebut the presumption that SEC regulation is likely to be better tailored than antitrust litigation to the particular features of the securities industry. (87)
a. Is the SEC Focused on Competition?
Although the SEC is required to take into account competition concerns when it promulgates rules, (88) two commentators recently suggested that the SEC is nevertheless unlikely to give competition values sufficient weight when making policy decisions. (89) The SEC, they reasoned, is "first and foremost an investor-protection and information-disclosure agency, not an agency that investigates and weeds out cartels or other anticompetitive practices." (90) Thus, they expressed skepticism that the SEC would adequately police competition. (91)
This argument is unconvincing for two reasons. First, it implies that regulation to promote competition in the securities industry and regulation to promote investor protection and information disclosure are disjoint sets. They are not. Most securities regulation designed to protect investors will in fact also promote competition and consumer welfare because investors are "consumers" of financial services. (92) Thus, activities that are injurious to competition and to consumer welfare will generally be inimical to the goal of investor protection as well. (93)
Second, to the extent that, as discussed above, competition and investor protection may at times be in tension, (94) there is no obvious reason to think that the SEC will exhibit systematic bias against competition. SEC personnel include economists and other policymakers who are capable of making analytical determinations concerning market concentration and power, barriers to entry, and the like. In addition, not only is the SEC required to consider competition effects when it promulgates regulations, but the Exchange Act expressly prohibits the SEC from promulgating a rule that will have a negative effect on competition unless it finds that such an effect is necessary or appropriate to pursue some other policy of the securities laws. (95) Just as important, the SEC is required to explain the likely competitive effects of a rule in the rule's published statement of basis and purpose. (96) This creates a focal point for a reviewing court considering a challenge to the rule under the Administrative Procedure Act (APA) as arbitrary and capricious or inconsistent with the Exchange Act's prohibition on unnecessary restraints on competition. (97) Moreover, in order to survive such review, the SEC will have to demonstrate that it considered public comments (98) received that were critical of the rule, as well as any academic studies that are inconsistent with the factual predicates of the rule, and will likely have to explain why it rejected possible alternative rules that might have been less injurious to competition. (99) Such "hard look" judicial review under the APA ensures that the SEC will not be able to blithely disregard its statutory mandate to regulate in the interests of competition, (100) even if for some reason it were otherwise inclined to do so. (101)
b. Agency Capture
A potentially more persuasive argument as to why the SEC might display excessive tolerance for anticompetitive activities in the securities industry is the agency capture theory, (102) which posits that regulated entities and organized special interests may exert undue influence on federal agencies. (103) Agency capture could manifest itself in a number of different forms. There may be a revolving door between the regulated industry and the agency such that current agency officials are reluctant to promulgate regulations that might upset their potential future employers. (104) More subtly, regulated firms may be more likely to participate in the notice-and-comment rulemaking process, (105) and thus their views may have a disproportionate effect on the form and content of the regulations ultimately adopted by the SEC. (106) Excessive industry influence is particularly likely to occur, it is said, when the costs of a proposed regulation affect a small concentrated group while the benefits from the regulation accrue to the public generally, as in the case of competition issues. (107) Because federal courts are presumably better insulated from the influence of the financial services industry than is the SEC, they might be thought to offer a superior venue for regulating competition even in the complex securities area. (108)
Commentators hold different views regarding the degree to which the SEC is likely to be subject to capture by the securities industry, (109) and a full examination of the issue is beyond the scope of this Note. However, several points bear emphasizing here, two of which have already been discussed. First, the agency capture concern is mitigated by the availability of judicial review of SEC regulations under the APA. (110) In addition, the Supreme Court recently held that even decisions not to initiate a rulemaking process are presumptively subject to review, albeit highly deferential review. (111) Thus, an unreasonable refusal by the SEC to prohibit certain anticompetitive conduct can potentially be challenged in an action brought under the APA. Moreover, to the extent that APA review of agency inaction may not be particularly rigorous, this approach seems consistent with Congress's intent to delegate broad rulemaking authority to the SEC through the securities laws. (112)
Second, and more fundamentally, even the weakest form of the implied antitrust immunity doctrine as predicated on SEC approval of a particular activity (113) contains an implicit premise, which is that Congress believes that the possibility of regulatory capture is outweighed by the SEC's comparative advantage over antitrust courts. A corollary is that if regulatory capture of the SEC is indeed a real problem in need of fixing, then denial of implied antitrust immunity seems to be a particularly blunt instrument with which to fix it. It is important to remember that the implied immunity battleground in the securities context is relatively narrow; the principal issue that critics of Billing have raised is whether, in the absence of SEC approval of a particular activity, the possibility that the SEC's position might change or that a court might misinterpret an SEC regulation is a sufficient basis for finding an implied repeal of the antitrust laws. (114) Egregious cases of agency capture are not going to be solved by adopting a narrow antitrust immunity doctrine precisely because the SEC can simply decide to issue an unambiguous blanket authorization for a particular activity in order to preclude antitrust suits. Rather, the appropriate forum in which to combat agency capture is judicial review under the APA. If such review is considered inadequate, the proper solution is for Congress to draft more pointed and narrowly worded statutes, not for courts to fail to give effect to broad delegations of power to the SEC by contracting the implied immunity doctrine.
A third reason that the regulatory capture argument is insufficient to overcome the presumption in favor of SEC regulation over antitrust litigation is that in the implied immunity context, a broad immunity doctrine can actually have the prophylactic, self-cleansing effect of reducing regulatory capture. This is because a broad immunity grant prevents the SEC from sharing responsibility for bad decisions with the antitrust courts. In the absence of a broad immunity standard, if Congress, the President, or the public perceives an overly lax antitrust policy, they may be unsure whether the SEC or the courts are responsible. By contrast, if silence or even an expression of disapproval of a particular activity by the SEC is enough impliedly to repeal the antitrust laws, then the SEC's failure nonetheless to take enforcement action against clearly anticompetitive and socially inefficient conduct will mean that Congress, the President, and the public will know whom to blame. Ex ante, this may lead the SEC to internalize more of the costs of its bad decisions. In this manner, giving the SEC more exclusive authority may actually reduce agency capture to the extent that the SEC will be punished to a greater degree if it behaves in a way that exhibits such capture.
c. The Unavailability of Treble Damages
Potentially the most rhetorically appealing objection to the argument that the SEC is likely to have an advantage over antitrust courts in distinguishing securities market activities that should be prohibited from those that should be allowed is that, even if the SEC possesses such an institutional advantage, it may not have the resources necessary to detect all harmful antitrust violations in the securities area. (115) In particular, the Clayton Act provides for treble damages against persons in violation of the antitrust laws. (116) By contrast, the SEC has no power to issue such a punitive remedy. The traditional justification for treble damages is that they are necessary to counterbalance underdetection and thereby force antitrust violators to internalize the full social costs of their actions. (117) Thus, one might expect that, without parallel antitrust actions, SEC regulation would lead to underdeterrence.
However, this argument ultimately rings hollow because the need for treble damages is likely to be fairly dubious in situations in which implied immunity might be found to exist. As scholars have recognized, the utility of treble damages as a mechanism for achieving optimal deterrence is positively related to the concealability of the antitrust offense at issue. (118) When a particular activity is relatively visible, the probability of detection may be high and extracompensatory damages may lead to overdeterrence rather than to optimal deterrence. In the context of implied immunity under the antitrust laws, the critical point is that a finding of immunity, even under the Billing standard, is predicated first and foremost on the SEC actually possessing jurisdiction over a particular activity and being actively engaged in the process of evaluating the costs and benefits of that activity. (119) But if an activity is such that it has captured the attention of regulators at the SEC in this way, then it is afortiori unlikely to be sufficiently concealable to justify a treble damages award. (120) By contrast, when a particular activity is difficult to detect and the SEC is thus unlikely to be aware of its existence or prevalence, a requisite condition for a finding of implied immunity--the exercise of the SEC's supervisory authority--will be absent, and antitrust suits will not be precluded. (121)
Of course, it is not inconceivable that, even if the SEC is actively engaged in reviewing the potentially anticompetitive effects of a particular securities market activity and has made a determination that the activity should be prohibited, it may still fail to detect particular instances of that activity. But the SEC has at its disposal a number of disciplinary tools with which to punish violators, tools that are significantly more focused and precise in their application than is automatic trebling under the antitrust laws. For example, willful violations of the securities laws or SEC regulations can lead to criminal sanctions (122) and, if committed by a broker-dealer, may be grounds for the SEC to suspend the broker-dealer's registration. (123) These mechanisms are adequate safeguards against the risk of underdeterrence of antitrust violations in the securities industry.
Thus, the unavailability of treble damages in SEC enforcement actions and private securities lawsuits is not a compelling argument against a broad implied immunity standard. When the SEC actively oversees a particular activity, permitting parallel antitrust suits in which damage awards are automatically trebled is an exceedingly heavy-handed method of solving any supposed detection problem and is instead likely to deter beneficial conduct. (124) As such, the treble damages argument does not undermine, and in fact affirmatively reinforces, the comparative advantage of SEC regulation over antitrust suits.
In sum, this Section has argued that the SEC serves an adequate antitrust function and that, because of antitrust courts' relative lack of securities-related expertise and comparatively narrow focus on competition, permitting parallel antitrust actions can inhibit socially beneficial securities marketing activities. Antitrust courts are therefore likely to prohibit too much conduct in the absence of regulatory guidance from the SEC. But they are also likely to prohibit too much conduct when the SEC has promulgated a regulation governing the activities at issue because, as the Court explained in Billing, a fine line often separates activities that the SEC permits from those that it prohibits. (125) As a result, courts might reach the merits in cases challenging conduct that the SEC would actually sanction, and the preceding analysis suggests that there is a material risk that they might erroneously impose liability for such conduct.
B. The Excessive Liability Concern
Not only does the SEC possess a comparative advantage over antitrust courts in determining whether a particular activity should be prohibited, but antitrust courts can be expected to impose excessive liability even as to those activities that both the SEC and the antitrust laws condemn. This conclusion rests on two assumptions. The first assumption is that harmful antitrust violations are likely to be relatively rare in the securities arena because the securities markets are competitive. The second assumption is that the variance of jury awards in antitrust cases is likely to be higher than the variance of penalties imposed by the SEC. If both of these assumptions are valid, then antitrust courts can be expected to impose excessive liability relative to the social optimum. The reason is that large upward errors in damage calculations are unbounded, whereas downward errors are bounded from below, and hence truncated, by zero. This Section will defend these two assumptions and then explain why antitrust courts, if left to their own devices, will impose excessive damage awards and produce overdeterrence in the securities area.
1. The Competitive Nature of the Securities Markets
Harmful anticompetitive activity in the securities industry is likely to be relatively uncommon. The reason, as Professor Hovenkamp has explained, is that there are thousands of sellers of any given security, and securities are close substitutes for one another. As a result, the securities markets are close to perfectly competitive. (126) The critical distinction to recognize is between the market for a firm's products or services and the market for ownership of that firm. Even if two companies produce entirely different products, the securities that they issue are close substitutes for one another. The value of those securities is a function of the cash flows that they can be expected to generate, whether in the form of regular dividends or a liquidating payment in the event that the company is sold, and the risk properties of those expected cash flows. As a result, assuming that the efficient capital market hypothesis holds and stock prices reflect all publicly available information, (127) any two stocks are essentially substitutes: they differ only in the extent to which their expected returns co-vary with the market (that is, their market betas). (128) Moreover, because investors can adjust the contents of the securities in their portfolios and the relative weights of those securities in any number of ways to achieve their desired levels of risk, there is no reason to believe that any one security does not compete with thousands of others. (129) In short, "all stocks are in the same relevant market insofar as antitrust is concerned." (130)
Because the economic concerns that animate the antitrust laws depend on an assumption that particular firms possess market power, (131) the competitive nature of the securities markets makes it unlikely that conduct that might be considered anticompetitive in other settings is in fact harmful in the securities context. (132) As an example, consider the tying scheme at issue in Billing, in which underwriters allegedly insisted that IPO investors purchase additional less attractive non-IPO shares. The theoretical economic concern with tying is that a firm with market power in the tying product market may use that power to suppress competition in the tied product market. (133) In the IPO context, the concern would be that some purchasers might prefer to purchase the tied security from a different seller but may end up purchasing it from the underwriter in order to obtain the IPO share. (134) A compelling argument can be made that tying arrangements generally will not impose any incremental harm on competition even when the defendant possesses market power in the tying product market because the defendant can only exercise that market power once, and any profits generated through sales of the tied product will be offset by a decline in profits from the tying product. (135) But even accepting the premise that in certain scenarios tying arrangements may enable a firm to expand the scope of its market power into the tied product market, the IPO context is pretty clearly not such a scenario. The reason is that any supposed anticompetitive effects of tying in this context are unlikely to be significant since sellers presumably do not have market power in the tying product, at least to the extent that IPO shares are not considered an independent market distinct from the equity market generally. (136)
The upshot is that the universe of activities that can be expected to harm competition in the securities markets is relatively limited. Of course, the mere fact that harmful anticompetitive conduct is rare is not necessarily sufficient to show that antitrust courts will impose excessive liability. A proponent of antitrust litigation (and one that rejects the comparative expertise of the SEC) might suggest that courts will recognize that many activities that might significantly reduce competition in other contexts are unlikely to do so in the securities arena and will simply adjust their damage awards accordingly. (137) Rather, the conclusion that antitrust courts will on average impose excessive liability rests on an additional assumption, defended in the following Subsection, that the variance of damage awards is likely to be greater in antitrust suits than in SEC enforcement actions.
2. The Effect of the Variance of Antitrust Damage Awards
In Billing, the Supreme Court expressed concern that the application of antitrust law to securities market activities might result in different antitrust courts rendering inconsistent judgments based on similar fact patterns. (138) To a considerable extent, this concern stems from the presence of juries in antitrust cases. (139) A number of commentators have documented that damage awards imposed by juries typically display significant variability. (140) Evidence suggests that antitrust cases are no exception. (141) Comparatively, it seems clear that SEC sanctions will be significantly less variable and more consistent from one enforcement action to another. There are several good reasons for this assumption. First, the variability of damage awards is inversely related to the number of people charged with making the damages determination because a greater number of decisionmakers dilutes the influence of outliers. (142) Because the SEC's enforcement division is larger by orders of magnitude than the size of juries deciding parallel antitrust cases, SEC enforcement actions can be expected to have a lower variance. (143) More importantly, any given antitrust jury hears only a single case and thus is incapable of making comparisons across cases. By contrast, the SEC brings many enforcement actions and can therefore reduce the variability of the penalties it seeks. (144)
The greater variability of antitrust jury verdicts will not in itself produce overdeterrence as long as companies are risk neutral. However, when combined with the fact that the universe of anticompetitive conduct in the securities industry is presumptively narrow, the variability of jury awards will in fact skew the expected level of damages upward. The reason is that large downward errors in the damages calculation are truncated by zero, whereas upward errors are unbounded. Although this truncation problem applies to any damages calculation, (145) it is particularly problematic when there is a significant chance that the true level of harm associated with a particular activity is low--as is the case with respect to antitrust violations in the securities markets--because this increases the probability that the truncated nature of the distribution will actually have the effect of reducing large downward errors. This skewing effect will be particularly pronounced in antitrust suits because upward errors are automatically trebled; (146) by contrast, the penalties that the SEC can seek are subject to a statutory cap. (147) Thus, even when the securities laws and the antitrust laws both prohibit a given activity, the variance of antitrust jury awards relative to SEC sanctions will cause antitrust courts to impose excessive liability on average.
Several potential counterarguments need to be addressed briefly. First, it might be argued that the greater variability of antitrust jury awards will be reduced through settlements. (148) But this argument misses the point: the variability of antitrust damages, combined with the competitive nature of the securities markets, means that ex ante the expected level of damages will exceed the efficient level of damages. This will skew the settlement amount upwards as well, so there will still be overdeterrence. (149)
A second possible objection to the argument presented in this Section is that some securities actions, like antitrust suits, are decided by juries and thus may also be susceptible to the same variability problem. For example, the SEC can opt to bring enforcement actions in federal court rather than in an administrative proceeding before an administrative law judge. (150) Moreover, the level of damages awarded in a private securities lawsuit will typically be determined by a federal jury.
There are several responses to this argument. As to the possibility that the SEC might pursue civil penalties through a jury trial, such litigation is not the norm. The vast majority of SEC enforcement actions settle, (151) and even cases that do not settle typically proceed through an administrative adjudication and result in the issuance of a cease-and-desist order, (152) through which the SEC may require disgorgement of the defendant's ill-gotten gains. (153) The SEC can also seek civil penalties in administrative adjudications against securities firms that willfully violate the securities laws. (154) When the SEC does pursue civil penalties in a federal jury trial, it usually settles with the defendant, (155) and unlike a private antitrust plaintiff the SEC can be expected to settle for an amount that, in its view, achieves the optimal level of deterrence ex ante.
The point about private securities lawsuits is more persuasive but ultimately unconvincing. The difference between private antitrust actions and securities suits is that the latter will not necessarily be available whenever a regulated company engages in alleged anticompetitive conduct, even if that conduct appears to violate an SEC regulation. Indeed, the Court has interpreted the antifraud provisions of section 10(b) of the Exchange Act and Rule 10b-5, the principal source of private securities suits, to create a private right of action only to redress material misrepresentations or omissions. (156) As such, absent a failure to disclose, conduct that allegedly harms competition may not be actionable. (157) Therefore, the proportion of cases decided by nonexpert juries is likely to be greater, and the variability problem correspondingly more acute, with respect to cases brought under the antitrust laws than those brought under the securities laws (typically in an SEC enforcement action). (158)
Finally, it might seem counterintuitive to suggest that antitrust courts would impose excessive liability on securities firms that engage in conduct that the SEC itself prohibits. But this is just a manifestation of the generally accepted phenomenon that excessive damage awards can lead to overdeterrence and may induce defendants to undertake socially inefficient precautions that are costlier than the harms that they seek to prevent. (159) For example, securities firms may engage in costly monitoring activities to ensure that their employees do not violate certain procompetitive SEC regulations governing underwriters or broker-dealers. (160) Efficiency requires that firms undertake such precautionary monitoring only to the extent that doing so costs less than the reduction in harm that it produces. (161) But if antitrust courts can be expected to impose liability in excess of the true level of harm to competition from a given activity, then securities firms may incur monitoring costs that exceed the actual social costs that such monitoring prevents. In addition, if regulated firms are not completely certain about whether a particular activity is permissible under applicable SEC regulations, then the threat of being exposed to excessive damages in an antitrust suit may induce them to refrain from socially valuable activities that the SEC would in fact permit but that are within this zone of uncertainty. (162)
Thus, the overdeterrence effect of excessive antitrust damage awards is indeed a legitimate concern even when the SEC disapproves of the challenged conduct. This is distinct from the Court's concern that antitrust courts may incorrectly interpret an SEC regulation and prohibit conduct that the SEC would permit. (163) An antitrust court may also interpret an SEC regulation correctly but impose excessive damages for activities that both the antitrust laws and the securities laws condemn.
III. THE OPTIMAL IMPLIED IMMUNITY STANDARD
Accepting the premise, defended in Part II, that antitrust courts are likely to prohibit too much conduct and impose too much liability in the securities context, what are the implications for the optimal implied immunity standard? Consider three possible standards in descending order of breadth. The first and broadest is the active review standard under which antitrust suits are impliedly precluded whenever the SEC has jurisdiction over a particular activity and is actively engaged in reviewing the merits of that activity, regardless of whether the SEC has promulgated an applicable regulation. The second is the some regulation standard, which would preempt antitrust suits only when the SEC has jurisdiction over a particular activity and has promulgated a regulation concerning the activity, whether permitting or prohibiting it. The third and narrowest possible standard is the affirmative approval standard, according to which an antitrust suit is only precluded when the SEC has promulgated a regulation permitting the activity in question. Billing clearly rejected the affirmative approval standard proposed by a number of commentators. (164) Since the SEC had promulgated a regulation and had issued guidance covering the alleged tying and laddering arrangements at issue, (165) the Court did not have occasion to distinguish between the active review standard and the some regulation standard, both of which are consistent with the outcome of the case. However, the Court's decisions in both Gordon and NASD suggest that the SEC need not affirmatively promulgate a regulation to preclude application of the antitrust laws, (166) which supports the active review standard.
This Part argues that a broad implied immunity standard predicated on the SEC's jurisdiction over, and active review of, a particular activity is efficient. But whereas in Billing the Court justified its implied immunity analysis by reference to the chilling effects of erroneous antitrust judgments ex post, this Part shifts the focus to ex ante regulatory action by the SEC. It argues that, from an ex ante perspective, the principal concern with a narrower implied immunity doctrine is that it might distort the SEC's regulatory decisions. In particular, if the SEC has three regulatory choices--prohibit a class of conduct entirely, permit it entirely, or adopt a nuanced rule that permits some forms of the conduct but prohibits others--and if a nuanced approach is optimal but a blanket authorization is preferable to a complete prohibition, then under either a some regulation standard or an affirmative approval standard the SEC might opt to permit the conduct in its entirety simply in order to preempt antitrust suits. (167)
The SEC can be expected to choose such a second-best solution in two situations. First, the SEC might opt for such a rule if it believes that it will not have time to study the activity at issue before an antitrust suit is resolved and that an antitrust court, if left to its own devices, might prohibit too much conduct or impose excessive liability for antitrust violations. Second, the SEC might choose to permit the entire class of conduct if it believes that, even if it were able to adopt a nuanced rule in time, a court might misapply that rule and prohibit conduct that the SEC would permit or award excessive damages for activities that the SEC prohibits. In combination, these two scenarios, which are modeled in the following Sections, suggest that an active review standard is optimal because it enables the SEC to regulate without solicitude for the possibility of erroneous decisions in antitrust cases. (168)
A. The Effect of Time Constraints on SEC Regulation
This Section will address the first problem with adopting either a regime of some regulation or an affirmative approval implied immunity standard. The concern with each of these standards is that the SEC may be in the process of studying a particular activity but may not think it has sufficient time to fashion a complex regulation distinguishing permissible from impermissible behavior before an antitrust case is decided. (169) This possibility is not merely academic. Indeed, in Billing itself the SEC filed an amicus brief with the district court arguing for a finding of implied immunity on the ground that it was still "actively engaged in considering appropriate responses to the very same types of conduct that plaintiffs' allege" at the time of the suit. (170) Under both the some regulation and affirmative approval standards, implied immunity cannot be predicated on agency silence. As a result, the SEC might respond to such time pressures by opting for a simpler rule, and one that can be promulgated more quickly, permitting all forms of the activity in order to preclude an antitrust suit. To the extent that a more nuanced rule prohibiting only some forms of the activity is preferable to such a blanket authorization, either of these narrower implied immunity doctrines will produce a deadweight loss.
The effect of time constraints on the SEC's regulatory choices can be modeled as follows. Assume that the SEC can adopt one of three positions with respect to a potentially anticompetitive activity affecting the securities markets. It can choose to prohibit the activity entirely, to permit it entirely, or to attempt to develop a more complex approach that would prohibit some forms of the activity but permit others. The SEC's position with respect to IPO laddering arrangements provides a good example of such a nuanced approach. As discussed in Billing, the SEC prohibits underwriters from soliciting aftermarket orders during the IPO but allows them to ask customers about their longerterm plans to purchase additional shares. (171) This approach is designed to accommodate the competing policy goals of discouraging artificial inflation of post-IPO stock prices while permitting some level of price stabilization by allocating IPO shares to investors with longer holding periods. (172) Assume that such a nuanced approach is socially optimal but that as between the extremes of completely prohibiting or permitting all forms of the activity, the latter result is preferable.
In particular, let B represent the per-period net benefits from the optimal rule and let A represent the per-period net benefits from allowing all forms of the activity. Assume that a complete prohibition of the activity produces a net benefit that is normalized to zero. Thus, by assumption, B > A > o. I will begin by considering a one-period scenario and will then extend the model to two periods. At time zero (To) the SEC has three options. It can (1) adopt Regulation A (allow all conduct), (2) study the problem in the hope of discovering Regulation B (the optimal nuanced rule), or (3) prohibit the conduct entirely. Further, assume that the SEC is certain that it will he able to promulgate Regulation A before an antitrust suit against a securities firm is resolved at [T.sub.1] but that, because Regulation B will require more careful study and research, the SEC might not be able to promulgate Regulation B in time. In particular, let Q be the probability that the SEC can craft and implement Regulation B by [T.sub.1].
If the SEC promulgates Regulation A, then an antitrust suit will be precluded. In addition, for now assume that if the SEC succeeds in promulgating Regulation B, then the court will interpret the regulation correctly and thus will not reach the merits in any case challenging an activity that the SEC permits. However, if the SEC attempts to develop Regulation B but fails to promulgate it by [T.sub.1], then antitrust suits will not be precluded. In particular, assume that if the SEC attempts to promulgate Regulation B but is unable to do so, it will not have time to promulgate Regulation A before an antitrust suit is brought. Because courts are assumed to lack the institutional capacity to sua sponte draw the fine distinctions that the SEC, because of its expertise, might be able to build into Regulation B, in the event that a court is forced to decide an antitrust case on a clean slate its only two options will be to prohibit or permit the activity entirely. Assume that the probability that courts will prohibit the activity (thereby generating a net benefit of zero) is P. (173) P reflects not only the possibility that the defendant will lose on the merits but also the possibility that a court will rule in a manner adverse to the defendant on a motion to dismiss or a motion for summary judgment and the defendant then settles. Such unfavorable pretrial rulings are assumed to lead other similarly situated firms to refrain from the conduct at issue for fear of sanction. (174)
The question is under what conditions the SEC will opt at To for Regulation A, the second-best solution, in order to preempt an antitrust suit, instead of trying to enact Regulation B and running the risk that a court may decide the case unconstrained by any SEC rules. The answer is that the SEC will compare the expected benefits from Regulation A with the expected benefits from pursuing Regulation B. The benefits from Regulation A are simply A, because Regulation A can be promulgated with certainty. The benefits from pursuing Regulation B are a function of the probability that the
Commission successfully implements it on time (Q) and the probability that, in the event that it fails to do so, a court will prohibit the conduct at issue (P). These expected benefits can be expressed as follows:
Expected Benefits (Regulation B) = Q x B + (1 - Q) x (P x 0 + (1 - P) x A)
Thus, the SEC will opt for Regulation A whenever A > Q x B + (1 - Q) x ((1 P) x A). The effect of changes in these variables on the SEC's regulatory choice is intuitive. As Q, the probability that the SEC will be able to promulgate Regulation B in time, increases, the SEC becomes less likely to opt for Regulation A, which is essentially a form of insurance against the possibility that it will fail to promulgate Regulation B and that an antitrust court will prohibit the conduct. For the same reason, if P--the probability that an antitrust court will erroneously prohibit the activity in the event that the SEC does not promulgate a regulation--is low, then the SEC is more likely to attempt to promulgate Regulation B, since this implies that the downside to failure to preclude an antitrust suit is lower. And as the differential between the net benefits from Regulation A (A) and those from Regulation B (B) narrows, the SEC is more likely to opt for Regulation A than to run the risk of failing to enact Regulation B by the time an antitrust suit is brought at [T.sub.1] and thereby to leave open the possibility that a court may prohibit the activity, which is by assumption the worst outcome.
Under a broad active review implied immunity standard, the SEC's decision calculus is different. An antitrust suit will be impliedly precluded regardless of whether the SEC opts for Regulation A or instead studies the activity at issue in the hope of coming up with an optimal regulation between [T.sub.0] and [T.sub.1]. Thus, even if the SEC fails to promulgate Regulation B by [T.sub.1], the result will be that it does not bring any enforcement actions and the net benefits will be A, just as if it had chosen to permit the activity by enacting Regulation A in the first place. As a result, under an active review standard the SEC would always opt to pursue Regulation B over Regulation A. Mathematically, the expected benefits from attempting to promulgate Regulation B are strictly greater than those associated with Regulation A:
Expected Benefits (Regulation B) = Q x B + (1 - Q) x A > Q x A + (1 - Q) x A=A = Benefits (Regulation A)
Therefore, by introducing the possibility that a court may erroneously prohibit the conduct at issue if the SEC fails to enact a regulation preempting an antitrust suit, a narrow implied immunity doctrine might have the effect of inefficiently distorting the SEC's regulatory decision ex ante, leading it to forgo the possibility of pursuing an optimal regulatory standard.
This distortion is less severe but nevertheless still present when the model is extended from one to two periods. The only significant difference between a one-period model and a two-period model is that, in the latter, the SEC is more likely to pursue Regulation B at To because even if it fails to implement it in the first period, it will likely succeed in doing so during the second period. By contrast, if the SEC adopts Regulation A at To, then it may have difficulty adopting Regulation B by [T.sub.2]. The reason is that, as a matter of administrative law, it may be easier for an agency to promulgate an initial regulation than to switch from one regulation to another. (175) To simplify the analysis, assume that if the SEC promulgates Regulation A at To, then it will be stuck with that rule for both periods, but that if it pursues Regulation B at To, then even if it is unable to implement it by [T.sub.1] it will certainly be able to do so by [T.sub.2]. Assuming the same probabilities and per-period benefits described above, the respective benefits from adopting Regulation A and pursuing Regulation B at To are as follows (176):
Benefits (Regulation A) = A + A = 2A
Expected Benefits (Regulation B) = Q x (B+B)+(1 - Q) x [P x (0 + B) + (1 - P) x (A + B)] = 2Q x B + (1 - Q) x [P x B + (1 - P) x (A + B)].
Thus, the SEC will settle for Regulation A if A > Q x B + 1/2 x (1 - Q) x [P x B + (1 - P) x (A + B)]. Some algebraic manipulation shows that the SEC is more likely to pursue Regulation B in this model than in the one-period model because the alternative route--adopting Regulation A--commits the agency to the second-best solution for two periods instead of one. (177) But the SEC may nonetheless opt for Regulation A if the probability that it will fail to implement Regulation B during period one (1 - Q) and the probability (P) that a court will incorrectly prohibit the conduct in that event are sufficiently high. Thus, as in the one-period model, the possibility that a court may reach the wrong decision if the SEC does not promulgate a rule precluding an antitrust suit may force the SEC's hand and induce it to adopt a rule that is easy to implement but socially suboptimal.
The concern with a narrow implied immunity doctrine presented in this Section is somewhat related to Billing's recognition that the SEC's position on an issue may be subject to change. But whereas in Billing the Court aimed to show that the antitrust laws could produce a future inconsistency with SEC regulation even where the two regimes are at present in harmony, (178) the analysis presented above implicates a slightly different concern: that the SEC's policy choice may be unduly influenced in the first place by the specter of antitrust litigation. When the SEC regulatory process is subject to time constraints, an implied immunity standard conditioned on either some regulation or a regulation affirmatively approving the conduct at issue is problematic because it may induce the SEC to promulgate an inefficient regulation quickly in order to preempt potentially erroneous antitrust judgments.
B. The Ability of Courts To Interpret SEC Regulations Correctly
Even if time constraints had no effect on the SEC's regulatory options, an implied immunity standard predicated on the SEC's active review of the challenged conduct would still be optimal. In the absence of time constraints, the some regulation standard and the active review standard are essentially interchangeable because the SEC can, by assumption, promulgate any rule it chooses at any time. And an active review standard is strictly preferable to an affirmative approval standard because the latter may cause the SEC's regulatory choice to be influenced by its belief that, were it to adopt a complex and nuanced regulation distinguishing permissible from impermissible activities in a certain area, an antitrust court might have difficulty applying it correctly, or that, even if a court did apply it correctly, it might impose excessive liability for conduct that the SEC prohibits.
The first of these two possible concerns--that antitrust courts might misinterpret SEC regulations--played a fundamental role in Billing's incompatibility analysis. (179) As discussed in Part I, the Court's fear derived from the fact that a fine line often separates conduct that the SEC prohibits from conduct that it permits and that the same piece of evidence might be consistent with a determination that particular conduct falls on either side of the line. (180) The second concern--that an antitrust court might award excessive damages for conduct that violates SEC regulations--did not feature in the Court's reasoning but is justifiable based on the analysis presented in Section II.B. This Section shows that, to the extent that the SEC shares either of these concerns, it may decide to adopt a second-best solution by giving the activity at issue its unqualified approval in order to immunize it from an antitrust challenge.
As in the previous Section, assume that the SEC has three regulatory options: it can prohibit an activity entirely, permit it entirely (Regulation A), or adopt a more subtle and complex rule prohibiting some forms of the activity but permitting others (Regulation B). The net benefits from a prohibition are normalized to zero, but the benefits from Regulation B are B and the benefits from Regulation A are A, where B > A > o. Unlike in the prior Section, assume that the SEC is unconstrained by time and can thus promulgate any of the three regulations with certainty. Under an affirmative approval standard, if the SEC chooses to permit the activity entirely it will preclude all antitrust suits, but if it adopts Regulation B, then antitrust suits will only be precluded if the court concludes that the conduct at issue in a particular case is permissible under the SEC's regulation. If the court concludes that the conduct violates the SEC's regulation, then it will proceed to the merits of the antitrust claims.
For simplicity, assume that there are two ways that courts might evaluate the activity at issue. On the one hand, with probability P courts will correctly interpret the regulation and prohibit activities that the SEC would prohibit and impose the optimal level of damages against violators but dismiss claims challenging conduct that the SEC permits. On the other, however, with probability 1 - P courts will either misapply the regulation by prohibiting more conduct than the SEC seeks to condemn or apply the regulation correctly but impose excessive penalties for conduct that the SEC prohibits (in either case, resulting in inefficient overdeterrence). (181) Specifically, because permitting the entire class of conduct is by assumption preferable to prohibiting it, when courts do make such errors the average benefits are assumed to be K x A, where 1 > K > 0. The model does not account for the possibility of false negatives--that a court might permit conduct that the SEC would in fact prohibit--because unlike false positives, false negatives can be corrected by the SEC by bringing a subsequent enforcement action. Under these assumptions, the expected benefits from enacting Regulation B can be expressed as follows:
Expected Benefits (Regulation B) = P x B + (1 - P) x K x A
Thus, the SEC will opt for Regulation A over Regulation B whenever A > P x B + (1 - P) x K x A, which is more likely to be the case if the SEC believes that courts will misinterpret a complex regulation and prohibit socially beneficial conduct or impose excessive damages for violations of the regulation (that is, if P is low). In addition, as the differential between the benefits from Regulation A and Regulation B shrinks, the SEC is more likely to opt for the former.
As in the previous Section, the SEC's decision to opt for Regulation A is a direct consequence of assuming a narrow implied immunity standard in the model--in particular an affirmative approval standard. If instead antitrust suits were precluded regardless of whether the SEC opted for Regulation A or Regulation B, then the SEC would always choose to promulgate Regulation B and would apply that regulation in enforcement actions to distinguish conduct that has a sufficiently harmful effect on competition to merit prohibition from conduct that lacks such an effect. The SEC, moreover, would impose an optimal sanction against firms that violate the regulation.
This suggests that as a policy matter the Supreme Court was correct to factor the possibility that antitrust courts might misapply SEC regulations into its implied immunity analysis. However, the Court's conclusion that in the absence of a finding of immunity the prospect of excessive antitrust liability might have a chilling effect on legitimate socially beneficial activities in the securities industry is only part of the story. (182) The opposite result might occur if the SEC, in order to preclude antitrust actions, adopts an overly permissive regulation ex ante. Thus, contrary to what the critics of Billing suggest, (183) a narrow implied immunity doctrine might actually lead to too little antitrust enforcement in the securities area, rather than too much.
The arguments presented in this and the previous Section imply that of the three potential implied immunity standards discussed in the beginning of this Part, the broadest active review standard is optimal. The possibility that a court might misapply an SEC regulation or impose excessive liability even against firms that do violate SEC regulations suggests that such a standard is superior to an affirmative approval standard--the narrowest standard--even assuming away the possibility that the SEC may be subject to time constraints in choosing between different regulatory options. Moreover, when time constraints are factored into the SEC's regulatory calculus, it becomes apparent that an active review standard is preferable to both a some regulation standard and an affirmative approval standard, each of which would permit antitrust suits in the absence of an SEC regulation and thus might induce the SEC to promulgate a permissive regulation quickly in order to preempt potentially erroneous antitrust decisions. To be sure, the preference for an active review standard ultimately depends on the assumptions articulated and defended in Part II: that antitrust courts can be expected to prohibit too much conduct and impose too much liability in the securities context. But to the extent that these assumptions hold, this Part has shown that the broad implied immunity standard for which Gordon, NASD, and Billing stand is efficient and may actually lead to more antitrust enforcement than would the narrower standard advocated by a number of commentators.
This Note has provided a defense of the Supreme Court's decision in Credit Suisse Securities (USA) LLC v. Billing. It has argued that Billing's emphasis on whether the SEC possesses and exercises jurisdiction over a particular activity is consistent with relevant Supreme Court precedents, which belie the claim that implied immunity has been found only when the SEC explicitly sanctions an activity. Moreover, this Note has argued that Billing was correctly decided as a normative matter as well. The SEC is better positioned than antitrust courts to determine whether a particular activity merits prohibition because it possesses expertise in the securities area that generalist courts lack and because the strictures of antitrust law are not well suited to accommodate some of the policy goals that undergird the securities laws. In addition, the competitive nature of the securities industry and the greater variability of antitrust damage awards as compared to penalties imposed by the SEC imply that, on average, antitrust courts will impose excessive liability even as to activities that both the securities laws and the antitrust laws condemn. As such, a narrow implied immunity doctrine might lead to overdeterrence ex post and might actually cause the SEC to adopt overly permissive regulations ex ante solely for the purpose of preempting antitrust suits. The central accomplishment of Billing's broad implied immunity standard is that it enables the SEC to regulate outside of the shadow of the antitrust laws.
(1.) 551 U.S. 264 (2007).
(2.) Id. at 279-84.
(3.) Id. at 280-82.
(4.) Id. at 282-83.
(5.) See id. at 284.
(6.) See, e.g., Luitgard W. Lina Chambers, Note, A New Supreme Court Securities Jurisprudence? How Credit Suisse Securities (USA) LLC v. Billing Eviscerates the Role of Antitrust in Securities Law, Resurrects the Primary Jurisdiction Doctrine, and Leaves Retail Investors Out in the Cold, 44 HOUS. L. REV. 1131, 1159 (2007); Stacey Sheely Chubbuck, Note, Securities Law and Antitrust Law: Two Legal Titans Clash Before the United States Supreme Court in Credit Suisse Securities v. Billing, 62 OKLA. L. REV. 145, 161-62 (2009); Justin Lacour, Note, Unclear Repugnancy: Antitrust Immunity in Securities Markets After Credit Suisse Securities (USA) LLC v. Billing, 82 ST. JOHN'S L. REV. 1115, 1141-44 (2008).
(7.) See supra note 6.
(8.) See infra note 74 and accompanying text.
(9.) See infra Section I.C.
(10.) See infra notes 75-76 and accompanying text.
(11.) See infra Section II.A.
(12.) See Herbert Hovenkamp, Antitrust Violations in Securities Markets, 28 J. CORP. L. 607, 610 (2003).
(13.) See infra Subsection II.B.2.
(14.) See Chubbuck, supra note 6, at 160-62; Jacob L. Kahn, Note, From Borden to Billing: Identifying a Uniform Approach to Implied Antitrust Immunity from the Supreme Court's Precedents, 83 CHI.-KENT L. REV. 1439, 1473-74 (2008); Lacour, supra note 6, at 1145-48.
(15.) 373 U.S. 341 (1963).
(16.) Id. at 343-45.
(17.) Id. at 347.
(18.) Id. at 349.
(19.) Id. at 357 (quoting United States v. Borden Co., 308 U.S. 188,298 (1939)).
(20.) Id. at 352 (quoting 15 U.S.C. [section] 78f(d) (1958)).
(21.) Id. at 357-58.
(22.) Id. at 358.
(23.) Id. at 358 n.12.
(24.) 422 U.S. 659, 661 (1975).
(25.) Id. at 680.
(26.) Id. at 682.
(27.) Id. at 682-85.
(28.) Id. at 689.
(29.) Id. at 690.
(30.) 422 U.S. 694 (1975).
(31.) Id. at 70l-03.
(32.) Id. at 720-21.
(33.) Id. at 721.
(34.) Id. at 727.
(35.) Id. at 728.
(36.) See Dr. Miles Med. Co. v. John D. Park & Sons Co., 220 U.S. 373 (1911), overruled by Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 552 U.S. 877 (2007).
(37.) NASD, 422 U.S. at 729-30. The Court similarly affirmed dismissal of the alleged horizontal conspiracy among the NASD and its members to suppress the secondary market on the ground that the SEC's exercise of authority was "sufficiently pervasive to confer an implied immunity." Id. at 730.
(38.) Id. at 734.
(39.) Id. at 718 n.31, 728; see also id. at 747-48 (White, J., dissenting) (criticizing the Court for "on the one hand concluding that the selling practices under scrutiny here are essential to the working of the statutory scheme" while at the same time acknowledging that the SEC had requested that the NASD prohibit resale price maintenance).
(40.) Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 269 (2007).
(41.) Id. at 269-70.
(42.) Id. at 275-76.
(43.) Id. at 276-77.
(44.) Id. at 276.
(45.) Id. at 280-82.
(46.) Given the breadth of the Court's incompatibility analysis, the decisive factors in future implied immunity cases are likely to be those that relate to the extent of SEC authority and oversight (factors one, two, and four).
(47.) See Billing, 551 U.S. at 279; Regulation M, 17 C.F.R. [section] [section] 242.100-.105 (2009) (prohibiting participants in a securities distribution from inducing aftermarket transactions); Commission Guidance Regarding Prohibited Conduct in Connection with IPO Allocations, Securities Act Release No. 8565, Exchange Act Release No. 51,500, Investment Company Act Release No. 26,828, 70 Fed. Reg. 19,672 (Apr. 13, 2005) (interpreting Regulation M).
(48.) See Billing, 552 U.S. at 280; Amendments to Regulation M: Anti-Manipulation Rules Concerning Securities Offerings, Securities Act Release No. 8511, Exchange Act Release No. 50,831, Investment Company Act Release No. 26,691, 69 Fed. Reg. 75,774, 75,785 (Dec. 17, 2004).
(49.) Billing, 551 U.S. at 281.
(50.) See id. at 282.
(51.) Id. at 283.
(52.) See supra note 39 and accompanying text.
(53.) See supra note 29 and accompanying text.
(54.) See supra note 14 and accompanying text.
(55.) Billing, 551 U.S. at 271 (quoting Silver v. N.Y. Stock Exch., Inc., 373 U.S. 341, 357 (1963)); United States v. Nat'l Ass'n of Sec. Dealers, 422 U.S. 694, 719-20 (1975) (quoting Silver, 373 U.S. at 357); Gordon v. N.Y. Stock Exch., Inc., 422 U.S. 659, 685 (1975) (quoting Silver, 373 U.S. at 357); Silver, 373 U.S. at 357.
(56.) See, e.g., Chubbuck, supra note 6, at 150-54; Kahn, supra note 14, at 1444-50.
(57.) 371 U.S. 296, 298, 300 n.5 (1963).
(58.) Id. at 309.
(59.) Id. at 310.
(60.) Phonetele, Inc., v. Am. Tel. & Tel. Co., 664 F.2d 716, 727 (9th Cir. 1981); see also Verizon Commc'ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 422-12 (2004) (discussing the importance of the regulatory characteristics of the industry at issue in informing antitrust analysis).
(61.) See, e.g., Otter Tail Power Co. v. United States, 410 U.S. 366, 373-74 (1973) (rejecting implied immunity where the legislative history of the Federal Power Act belied a "purpose to insulate electric power companies from the operation of the antitrust laws"); United States v. Phila. Nat'l Bank, 374 U.S. 321, 352 (1963) (declining to imply antitrust immunity under the Bank Merger Act because the legislative history "seems clearly to refute any suggestion that applicability of the antitrust laws was to be affected"); United States v. Radio Corp. of Am., 358 U.S. 334, 346 (1959) (rejecting implied immunity because the legislative history of the Radio and Communications Acts demonstrated Congress's intent not to preempt antitrust enforcement).
(62.) Phonetele, 664 F.2d at 727 n.31.
(63.) Id. at 728.
(64.) See sources cited supra note 6.
(65.) See Chubbuck, supra note 6, at 161; Kahn, supra note 14, at 1473 n.290; Lacour, supra note 6, at 1119.
(66.) See supra notes 29, 39 and accompanying text.
(67.) See Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 273 (2007) (observing that in Gordon the Court declined to issue an injunction as to future rate fixing even though securities regulation and the antitrust laws would likely both prohibit rate fixing going forward).
(68.) See Kahn, supra note 14, at 1488; Lacour, supra note 6, at 1143.
(69.) See Billing, 551 U.S. at 176-77.
(70.) See Gordon v. N.Y. Stock Exch., Inc., 422 U.S. 659, 679-80 (1975).
(71.) United States v. Nat'l Ass'n of Sec. Dealers, Inc., 422 U.S. 694, 721 n.33 (1975) (quoting the Investment Company Act, 15 U.S.C. [section] 80a-22(f) (1970)).
(72.) See Cynthia A. Williams, The Securities and Exchange Commission and Corporate Social Transparency, 112 HARV. L. REV. 1197, 1235 n.199 (1999) (referring to the "pattern of broad delegation to the SEC found generally in the Exchange Act").
(73.) See generally Daniel A. Crane, Technocracy and Antitrust, 86 TEX. L. REV. 1159, 1185 (2008) (discussing the Court's explicit recognition of the risk of false positives in antitrust cases); Bruce H. Kobayashi & Joshua D. Wright, Federalism, Substantive Preemption, and Limits on Antitrust: An Application to Patent Holdup, 5 J. COMPETITION L. & ECON. 469, 472-74 (2009) (arguing that recent Supreme Court decisions have been motivated by concerns regarding the possibility of false positives by antitrust courts).
(74.) See Chubbuck, supra note 6, at 166; Kahn, supra note 14, at 1492; Lacour, supra note 6, at 1152-53.
(75.) See 15 U.S.C. [section] 77b(b) (2006) (providing that when the SEC engages in rulemaking pursuant to the Securities Act and "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"); id. [section] 78w(a)(2) ("The Commission and the Secretary of the Treasury, in making rules and regulations pursuant to any provisions of [the Exchange Act], shall consider among other matters the impact any such rule or regulation would have on competition. The Commission and the Secretary of the Treasury shall not adopt any such rule or regulation which would impose a burden on competition not necessary or appropriate in furtherance of the purposes of [the Exchange Act].").
(76.) Id. [section] 78w(a)(2).
(77.) See Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 279-81 (2007); Gordon v. N.Y. Stock Exch., Inc., 422 U.S. 659, 689-90 (1975) (referring to the "expertise of the SEC" and the "confidence the Congress has placed in the agency"); Hovenkamp, supra note 12, at 629 (arguing that regulatory agencies have greater expertise than antitrust courts).
(78.) See Nat'l Soc'y of Prof'l Eng'rs v. United States, 435 U.S. 679, 692 (1978) (explaining that "the purpose of [antitrust] analysis is to form a judgment about the competitive significance of the restraint; it is not to decide whether a policy favoring competition is in the public interest").
(79.) See Gordon, 422 U.S. at 689; Hovenkamp, supra note 12, at 609, 633; see also Paul S. Atkins & Bradley J. Bondi, Evaluating the Mission: A Critical Review of the History and Evolution of the SEC Enforcement Program, 13 FORDHAM J. CORP. & FIN. L. 367, 369 (2008) (discussing the "multidimensional nature of the SEC's mission").
(80.) The legislative history of the relevant amendments to the Exchange Act suggests that the SEC is not necessarily obligated to elevate competition concerns above all others. S. REP. NO. 94-75, at 13 (1975), reprinted in 1975 U.S.C.C.A.N. 179, 191; see also Brief for Petitioners at 10, Billing, 551 U.S. 264 (No. 05-1157) (observing that Congress specifically rejected a Justice Department proposal that would have required the SEC to adopt a "competition first" policy when promulgating regulations under the securities laws).
(81.) See Hovenkamp, supra note 12, at 609; John P. Lucas, Casenote, Pruning the Antitrust Tree: Credit Suisse Securities (USA) LLC v. Billing and the Immunization of the Securities Industry from Antitrust Liability, 59 MERCER L. REV. 803, 816 (2008).
(82.) See 15 U.S.C. [section] 78i(a)(6) (2006); 17 C.F.R. [section] 242.104 (2009).
(83.) See United States v. Socony-Vacuum Oil Co., 310 U.S. 150, 221 (1940).
(84.) Cf. Friedman v. Salomon/Smith Barney, Inc., No. 98 Civ. 5990 (NRB), 2000 WL 1804719, at * 11 (S.D.N.Y. Dec. 8, 2000) (finding implied immunity for antiflipping rules imposed by underwriters, which prohibit purchasers from selling IPO shares in the aftermarket within a specified time period, based on the SEC's "studied assessment that the benefits of price stabilization to the capital markets outweigh the admitted anti-competitive aspects"), aff'd, 313 F.3d 796 (2d Cir. 2002).
(85.) See Billing, 551 U.S. at 279 (noting that respondents "must concede antitrust immunity" for activities which the SEC permits or encourages).
(86.) See Memorandum Amicus Curiae of the SEC at 19, In re Initial Pub. Offering Antitrust Litig., 287 F. Supp. 2d 497 (S.D.N.Y. 2003) (No. 01 Civ. 2014) (arguing that by directing the SEC to "balance competition in its rule makings," Congress has made manifest its intention to preempt antitrust suits).
(87.) In the last several decades, the Supreme Court has abandoned per se antitrust rules in favor of a rule-of-reason analysis for a number of practices. See, e.g., Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877 (2007) (applying the rule of reason to resale price maintenance); NW. Wholesale Stationers, Inc. v. Pac. Stationery & Printing Co., 472 U.S. 284, 296-97 (1985) (limiting the application of the per se rule against group boycotts to cases in which the defendants possess market power) ; Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 16-18 (1984) (restricting the per se rule against tying to cases in which the defendant possesses market power in the tying product); Cont'l T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36 (1977) (applying the rule of reason to vertical territorial restrictions). Some commentators have argued that the rule of reason enables antitrust courts to adapt their analyses to the particular concerns of the securities industry and thus undermines the assumption that courts will prohibit too much conduct. See Stacey L. Dogan & Mark A. Lemley, Antitrust Law and Regulatory Gaming, 87 TEX. L. REV. 685, 700-01 (2009); Kahn, supra note 14, at 1494. But this argument is unconvincing since it is still the case that nonexpert and generalist judges and juries are applying the rule-of-reason analysis, and the rule of reason still focuses narrowly on competition to the exclusion of other policy goals. Nat'l Soc'y of Prof'l Eng'rs v. United States, 435 U.S. 679, 691-92 (1978).
(88.) See supra note 75 and accompanying text.
(89.) Dogan & Lemley, supra note 87, at 697-98.
(90.) Id. at 698.
(92.) See Howell E. Jackson, A System of Selective Substitute Compliance, 48 HAR v. INT'L L.J. 105, 112 (2007) (using the terms "investors" and "consumers" interchangeably and noting that enhanced competition among exchanges and broker-dealers works to their benefit).
(93.) See, e.g., Gordon v. N.Y. Stock Exch., Inc., 422 U.S. 659, 674-76 (1975) (explaining that the SEC's decision to disapprove of fixed commission rates was motivated by investor protection concerns).
(94.) See supra Subsection II.A.1.
(95.) See 15 U.S.C. [section] 78w(a)(2) (2006).
(97.) See 5 U.S.C. [section][section] 706(2)(A), (C) (2006).
(98.) See id. [section] 553(c).
(99.) See Motor Vehicle Mfrs. Ass'n of the U.S.v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (1983) (discussing the nature of arbitrary and capricious review).
(100.) See, e.g., Thomas J. Miles & Cass R. Sunstein, The Real World of Arbitrariness Review, 75 U. CHI. L. REV. 761, 761-65 (2008) (discussing the practical effects of hard look review).
(101.) The SEC might be able to avoid the rigors of hard look judicial review by relying on general policy statements or interpretive rules, which, unlike legislative rules, are exempt from the APA's notice-and-comment requirements. See 5 U.S.C. [section] 553(b) (3) (A). However, issuing guidance documents cannot completely displace legislative rulemaking since such documents do not carry the force of law, and if the SEC attempts to treat them otherwise a court may invalidate them. See Am. Mining Cong. v. Mine Safety & Health Admin., 995 F.2d 1106, 1112 (D.C. Cir. 1993); Pac. Gas & Elec. Co. v. Fed. Power Comm'n, 506 F.2d 33, 39 (D.C. Cir. 1974).
(102.) Some seminal works in this area include JAMES M. BUCHANAN & GORDON TULLOCK, THE CALCULUS OF CONSENT: LOGICAL FOUNDATIONS OF CONSTITUTIONAL DEMOCRACY (1962); DANIEL A. FARBER & PHILIP P. FRICKEY, LAW AND PUBLIC CHOICE: A CRITICAL INTRODUCTION (1991); MANCUR OLSON, THE LOGIC OF COLLECTIVE ACTION: PUBLIC GOODS AND THE THEORY OF GROUPS (1965); and Richard A. Posner, Theories of Economic Regulation, 5 BELL J. ECON. & MGMT. SCI. 335 (1974).
(103.) See Elena Kagan, Presidential Administration, 114 HARV. L. REV. 29.45, 2264-65 (2001).
(104.) See, e.g., John M. Griesbach, Some Observations About the Turn Toward Federal Rulemaking in Health Law, 49 ST. LOUIS U. L.J. 141, 146 (2004).
(105.) See Dogan & Lemley, supra note 87, at 699.
(106.) See Kagan, supra note 103, at 2265.
(107.) See Dogan & Lemley, supra note 87, at 699.
(108.) Id. at 699-700.
(109.) Compare Francis J. Facciolo, When Deference Becomes Abdication: Immunizing Widespread Broker-Dealer Practices from Judicial Review Through the Possibility of SEC Oversight, 73 MISS. L.J. 1, 6 (2003) (suggesting that "the SEC is subject to heavy political pressures when it adopts a position unpopular with the financial services industry"), with Herbert Hovenkamp, Antitrust and the Regulatory Enterprise, 2004 COLUM. BUS. L. REV. 335, 359 (arguing that public choice theory lacks "empirical robustness"), and Hovenkamp, supra note 12, at 629 ("[T]here is little reason for thinking that the SEC is being so solicitous of the market participants under its jurisdiction that it is ignoring the competitive implications of practices that it approves.").
(110.) See supra notes 97-101 and accompanying text.
(111.) Massachusetts v. EPA, 549 U.S. 497 (2007).
(112.) See supra note 72 and accompanying text.
(113.) This has never been an accurate statement of the doctrine. See supra Part I.
(114.) See supra Part I.
(115.) See Louis L. Jaffe, Primary Jurisdiction, 77 HARV. L. REV. 1037, 1053 (1964) ("[A] federal agency's resources of money, time, and men are rarely adequate to its farflung jurisdiction ...."); Chambers, supra note 6, at 1163 (suggesting that the expertise of regulatory agencies might be "drained through plain underfunding").
(116.) 15 U.S.C. [section] 15(a) (2006).
(117.) See Gary S. Becker, Crime and Punishment: An Economic Approach, 76 J. POL. ECON. 169, 199 n.55 (1968); Peter G. Bryant & E. Woodrow Eckard, Price Fixing: The Probability of Getting Caught, 73 REV. ECON. & STAT. 531, 535 (1991).
(118.) See, e.g., Frank H. Easterbrook, Detrebling Antitrust Damages, 28 J.L. & ECON. 445, 449, 455 (1985).
(119.) Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 276-77 (2007).
(120.) Cf. Hovenkamp, supra note 12, at 629 (suggesting that SEC regulation is an adequate substitute for treble damages in terms of deterrence because "[a]gency regulation is more likely to occur before-the-fact through rule-making or reporting").
(121.) See Billing, 551 U.S. at 275-76.
(122.) See 15 U.S.C. [section][section] 77x, 78ff(a), 80a-48, 80b-7 (2006).
(123.) See id. [section] 780(b)(4).
(124.) See infra notes 159-162 and accompanying text.
(125.) See supra notes 46-50 and accompanying text.
(126.) See Hovenkamp, supra note 12, at 610. This analysis would not apply if one viewed IPO stocks and stocks of established issuers as nonfungible. In addition, this analysis does not apply to collusive agreements among underwriters with respect to fees charged to issuers, as distinguished from investors, since the market for underwriters may be sufficiently concentrated that individual underwriters possess market power vis-a-vis issuers. See id. at 618.
(127.) See JAMES D. COX, ROBERT W. HILLMAN & DONALD C. LANGEVOORT, SECURITIES REGULATION: CASES AND MATERIALS 105-08 (5th ed. 2006).
(128.) See RICHARD A. BREALEY, STEWART C. MYERS & FRANKLIN ALLEN, PRINCIPLES OF CORPORATE FINANCE 188-91 (8th ed. 2006).
(129.) See Merritt B. Fox, Civil Liability and Mandatory Disclosure, l09 COLUM. L. REV. 237, 276 n.98 (2009) ("[A] vast reservoir of other stocks with the same beta could substitute for the issuer's shares."); Ronald J. Gilson & Reinier H. Kraakman, The Mechanisms of Market Efficiency, 70 VA. L. REV. 549, 570 n.67 (1984) (describing securities as "fungible commodities with large numbers of near-perfect risk-return substitutes").
(130.) Hovenkamp, supra note 12, at 612.
(131.) See Frank H. Easterbrook, The Limits of Antitrust, 63 TEX. L. REV. 1, 20 (1984) ("Firms that lack power cannot injure competition no matter how hard they try.").
(132.) See Hovenkamp, supra note 12, at 611-12.
(133.) See David F. Shores, Economic Formalism in Antitrust Decisionmaking, 68 ALB. L. REV. 1053, 1071 (2005).
(134.) See Hovenkamp, supra note 12, at 627-28.
(135.) See Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2, 36 (1984) (O'Connor, J., concurring) ; ROBERT H. BORK, THE ANTITRUST PARADOX: A POLICY AT WAR WITH ITSELF 373 (1978).
(136.) See Hovenkamp, supra note 12, at 628; see also Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 286 (2007) (Stevens, J., concurring) (suggesting that 1PO stocks compete with "the vast multitude of other securities traded in a free market").
(137.) On the other hand, depending on the nature of the alleged restraint, courts may not even reach the question of whether the defendant possesses market power. See Nat'l Collegiate Athletic Ass'n v. Bd. of Regents of the Univ. of Okla., 468 U.S. 85, 109 (1984) ("As a matter of law, the absence of proof of market power does not justify a naked restriction on price or output....").
(138.) Billing, 551 U.S. at 281.
(139.) See id.; Hovenkamp, supra note 12, at 629.
(140.) See, e.g., Shari Seidman Diamond, Michael J. Saks & Stephan Landsman, Juror Judgments About Liability and Damages: Sources of Variability and Ways To Increase Consistency, 48 DEPAUL L. REV. 301, 315 (1998) (finding "a substantial potential for variability in jury awards based merely on the make-up of the particular jury that decides a case"); Michael J. Saks, If There Be a Crisis, How Shall We Know It?, 46 MD. L. REV. 63, 76 (1986) ("Some data do suggest that jury awards have grown increasingly unpredictable."); Frederick N. Smalkin & Frederick N.C. Smalkin, The Market for Justice, the "Litigation Explosion" and the "Verdict Bubble": A Closer Look at Vanishing Trials, 1 FED. COURTS L. REV. 417, 430 (2006) ; Michael J. Saks, Blaming the Jury, 75 GEO. L.J. 693, 710 n.62 (1986) (reviewing VALERIE P. HANS & NElL VIDMAR, JUDGING THE JURY (1986)) (referring to the increased variability and decreased predictability of jury awards).
(141.) See Shari Seidman Diamond & Jonathan D. Casper, Blindfolding the Jury to Verdict Consequences: Damages, Experts, and the Civil Jury, 26 LAw & SOC'Y REV. 513, 530-31 (1992) (empirically studying jurors' damage awards in a hypothetical price-fixing case and finding that "award size ... is characterized by high variance") ; see also id. at 547-53 (finding that the damage estimation of the jury foreperson tends to have a disproportionate influence on the level of damages ultimately determined by the jury, which again suggests variability from one jury to another based merely on the identity of the juries' respective forepersons).
(142.) See Diamond et al., supra note 140, at 317-18.
(143.) Even though the staffing of any given enforcement action is likely to vary, the Commissioners will be involved in any proceeding and can solicit the views of other agency personnel, thereby minimizing the impact of outliers.
(144.) Cf. Roselle L. Wissler, Allen J. Hart & Michael J. Saks, Decisionmaking About General Damages: A Comparison of Jurors, Judges, and Lawyers, 98 MICH. L. REV. 751, 794 (1999) (finding that, in the tort context, "the variation of responses of different people to the same injury was greater for jurors than for ... other groups" such as lawyers and judges because "jurors have essentially no experience assigning a dollar value to injuries while the other groups do").
(145.) For a similar argument in the bankruptcy context, see Alan Schwartz, A Normative Theory of Business Bankruptcy, 91 VA. L. REV. 1199, 1232-33 (2005).
(146.) 15 U.S.C. [section] 15a (2006).
(147.) See id. [section][section] 77t(d)(2), 78u(d)(3), 78u-2(b), 80a-41(e), 80b-9(e).
(148.) See Crane, supra note 73, at 1184 (observing that most antitrust cases settle or are dismissed on a motion to dismiss or through summary judgment).
(149.) Courts frequently do not take into account the automatic trebling of antitrust damage awards when determining whether a settlement offer in a class action is adequate. Christopher R. Leslie, De Facto Detrebling: The Rush to Settlement in Antitrust Class Action Litigation, 50 ARIZ. L. REV. 1009, 1018 (2008). However, as explained above, treble damages are inappropriate in cases that might fit within Billing's implied immunity standard because the presence of SEC supervision makes it less likely that violations will remain undetected. See supra notes 119-120 and accompanying text. Thus, even if settlements systematically failed to account for the possibility of treble damages, such a failure would not lead to underdeterrence but instead would simply mitigate a different source of overdeterrence--treble damages themselves. Moreover, the fact that courts may not consider treble damages in ruling on the adequacy of a class action settlement does not mean that lead plaintiffs and their lawyers are unable to use the trebling feature as a bargaining chip in negotiating a higher settlement in the first place.
(150.) See 15 U.S.C. [section][section] 77t(b), 77t(d), 78u(d), 80a-41(d) to -41(e), 80b-9(d) to -9(e).
(151.) Cox ET AL., supra note 127, at 803 (noting that over ninety percent of SEC enforcement proceedings settle).
(152.) See id. at 812. The SEC is authorized to issue cease-and-desist orders for violations of the Securities Act or the Exchange Act, as well as regulations promulgated thereunder, by 15 U.S.C. [section][section] 77h-1, 78u-3.
(153.) See id. [section][section] 77h-1(e), 78u-3(e).
(154.) See id. [section] 78u-2; see also Arthur B. Laby & W. Hardy Callcott, Patterns of SEC Enforcement Under the 1990 Remedies Act: Civil Money Penalties, 58 ALB. L. REV. 5, 5-6 (1994) (discussing the SEC's enforcement tools).
(155.) See id. at 21 ("The vast majority of district court cases in which the SEC has sought penalties have settled.").
(156.) See Santa Fe Indus., Inc. v. Green, 430 U.S. 462, 474-77 (1977).
(157.) In addition, as the Court observed in Billing, Congress has erected heightened procedural requirements applicable to securities fraud actions. See Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 284 (2007) (discussing the Private Securities Litigation Reform Act of 1995, Pub. L. No. 104-67, 109 Stat. 737).
(158.) Certain Supreme Court decisions have made it more difficult for antitrust suits to proceed to trial, either by imposing more stringent pleading requirements on plaintiffs, see, e.g., Bell Atl. Corp. v. Twomhly, 550 U.S. 544, 556 (2007); Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 587-88 (1986), or by restricting the class of plaintiffs who have standing to sue, see, e.g., Ill. Brick Co. v. Illinois, 431 U.S. 720, 745-47 (1977); Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977). Although these legal developments mean that fewer antitrust cases will reach the jury, there is no reason to think that they will reduce the variance of jury awards among those cases that do make it to trial.
(159.) See A. Mitchell Polinsky & Steven Shavell, Punitive Damages: An Economic Analysis, 111 HARV. L. REV. 869, 879 (1998) (observing that "if damages exceed harm, firms might be led to take socially excessive precautions").
(160.) See id. (noting that an important type of precaution is "the variety of ways in which firms monitor and screen their employees").
(161.) See id. at 879-80.
(162.) See John E. Calfee & Richard Craswell, Some Effects of Uncertainty on Compliance with Legal Standards, 70 VA. L. REV. 965, 986 (1984) ("[E]xcessive damage awards will tend to increase any incentives to overcomply...."); Polinsky & Shavell, supra note 159, at 884-85 (noting that when there is a chance that parties may inaccurately assess the applicable legal standard they may take excessive precautions and that "raising the level of damages imposed on them will only exacerbate this problem").
(163.) Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 282 (2007).
(164.) See sources cited supra note 6.
(165.) See supra notes 47-48 and accompanying text.
(166.) See supra notes 25, 33 and accompanying text.
(167.) Because the SEC did in fact adopt a fairly nuanced approach to the laddering and tying arrangements at issue in Billing, the arguments presented below might seem inapplicable to the facts of the case. But the SEC presumably expected that antitrust actions would be preempted given the precedents discussed in Part I. Thus, it might in fact be precisely because of the Court's broad implied immunity doctrine that the SEC was able to issue finely drawn guidance with respect to the conduct challenged in Billing.
(168.) In practice, the SEC would not justify its regulatory choices by reference to the possibility of erroneous antitrust decisions. Nevertheless, such concerns might have a subtle and even unacknowledged influence on the form of regulation ultimately adopted.
(169.) See Hovenkamp, supra note 109, at 346 (arguing that agency indecisiveness should not be a bar to immunity because "[i]f a group of experts studying an issue for many years could not come to a clear understanding, then it would be imprudent to permit the issue to be decided by a jury trial in an antitrust case" and also that "[d]oing so would render agency regulation moot on that point"); Hovenkamp, supra note 12, at 632-33 (arguing that if the agency is continuously reviewing an activity but has not yet reached a determination as to its permissibility, there should be implied immunity).
(170.) Memorandum Amicus Curiae of the SEC at 32, In re Initial Pub. Offering Antitrust Litig., 287 F. Supp. 2d497 (S.D.N.Y. 2003) (No. 01 Civ. 2014).
(171.) See Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 280-81 (2007).
(172.) See Brief for Petitioners at 4-5, Billing, 551 U.S. 264 (No. 05-1157).
(173.) The model can also be interpreted to reflect the possibility that an antitrust court might impose excessive liability for conduct that does merit prohibition. See supra Section II.B. Assuming that the social costs of such excessive liability are roughly the same as the costs of false positives by courts, P can be conceptualized to reflect both the probability of such false positives and the probability of excessive damage awards.
(174.) See, e.g., Kobayashi & Wright, supra note 73, at 473 (observing that "false positives in antitrust ... are felt by many firms when some conduct is condemned"); Kahn, supra note 14, at 1480 ("[A] single conflicting judgment has the potential to affect the conduct of every regulated entity in the industry.").
(175.) For example, under the well-known Chevron doctrine, courts will defer to an agency regulation interpreting its substantive statute as long as the statute is ambiguous and the agency's construction is reasonable. Chevron U.S.A. Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 843-44 (1984). However, an agency's decision to move from one permissible interpretation to another will be subject to hard look review. See Motor Vehicle Mfrs. Ass'n of the U.S. v. State Farm Ins. Co., 463 U.S. 29 (1983) (finding an agency's decision to rescind a prior rule to be arbitrary and capricious under the APA even though the original rule and the absence of the rule both conformed with the substantive statute at issue). But see FCC v. Fox Television Stations, Inc., 129 S. Ct. 1800, 1810, 1811 (2009) (explaining that an agency's policy change is not subject to "more searching review" than the initial adoption of the policy but acknowledging that in some cases where the factual findings underlying the new policy contradict those underlying the old policy or where there are reliance interests at stake an agency must "provide a more detailed justification than what would suffice for a new policy created on a blank slate").
(176.) For simplification, I assume that the applicable discount rate is zero.
(177.) In the one-period model, the threshold level above which the SEC will opt for Regulation A is Q x B + (1 - Q) x (P x 0 + (1 - P) x A), which is strictly less than the threshold level in the two-period model, Q x B + 1/2 x (1 - Q) x [P x B + (1 - P) x (A + B)], because (1 - Q) x (P x 0 + (1 - P) x A) < 1/2 x (1 - Q) x [p x B + (1 - P) x (A + B)]. This last step follows from dividing each side of the inequality by (1 - Q) and simplifying. The right side becomes 1/2 x [(1 - P) x A + B], which is greater than (1 - P) x A because B > A > (1 - P) x A.
(178.) See Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 280-81 (2007).
(179.) See supra notes 47-49 and accompanying text.
(180.) See Billing, 551 U.S. at 281; supra Section I.B.
(181.) See supra text accompanying notes 159-163.
(182.) Billing, 551 U.S. at 282.
(183.) See supra note 6 and accompanying text.
Yale Law School, J.D. 2010; Brown University, A.B. 2007. I thank George Priest for providing helpful comments on an earlier draft. I am also grateful to Julia Malkina for superb editorial assistance, as well as to Alexandra Briggs and Ben Klein for their insights and feedback.
|Printer friendly Cite/link Email Feedback|
|Author:||Kling, Jacob A.|
|Publication:||Yale Law Journal|
|Date:||Jan 1, 2011|
|Previous Article:||Defining family in immigration law: accounting for nontraditional families in citizenship by descent.|
|Next Article:||Trade secret law and the changing role of judge and jury.|