Biff Campbell thinks Reg D has failed its intended purpose and the reason is state law.  Here’s part of the abstract:

Regulation D * * * offers businesses — especially businesses with relatively small capital requirements — fair and efficient access to vital, external capital.  * * * The data show that Regulation D is not working in the way the Commission intended or in a way that benefits society. The data reveal that companies attempting to raise relatively small amounts of capital under Regulation D overwhelmingly forego the low transaction costs of offerings under Rule 504 and Rule 505 in favor of meeting the more onerous (and more expensive) requirements of Rule 506. Additionally, these companies overwhelmingly limit their relatively small offerings to accredited investors, which dramatically reduces the pool of potential investors. This unintended and bad outcome is the result of the burdens imposed by state blue sky laws and regulations, and this has to a large degree wrecked the sensible and balanced approach of the Commission in Regulation D.  Congress. . . could solve the problem by expanding federal preemption to cover all offerings made under Regulation D.

He has a point.  Permitting the states to regulate national securities transactions enables individual states to impose regulatory costs outside their borders for the benefit of local interest groups.  This can have perverse effects — in this case, by letting individual states impede national capital formation and entrepreneurship .Indeed, a key economic rationale for federal law is to address this problem.  See Easterbrook & Fischel, Mandatory Disclosure for the Protection of Investors, 70 Virginia Law Review 669 (1984).  

But we don’t have to eliminate state securities laws, along with state law’s potential advantages of competition and experimentation, to deal with this problem.  There’s an alternative:  apply state law only to intrastate transactions, or to corporations that have contracted for the securities law of a particular state (e.g., by incorporating in the state).  In other words, apply the same choice-of-law rule to state securities law as to state corporate governance law.  I discuss this approach in Dabit, Preemption and Choice of Law and Preemption and Choice-of-Law Coordination (with O’Connor).

Ted Frank responds that the fact that the company had settled criminal allegations doesn’t mean the board was irresponsible given big companies’ exposure to prosecutorial overreaching (here’s my thoughts on the problems with prosecutors).  He notes that given huge potential penalties and legal costs “even a risk-neutral set of executives would refuse to go to trial on criminal charges that they had a 95% chance of winning.”  As Ted says:

The issue is this: first, any corporate law is going to have to balance false negatives (valid suits against directors being thrown out prematurely) and false positives (invalid suits against directors costing tens of millions of dollars in time and money to resolve). Any opening up of the courtroom doors to challenge directors will reduce false negatives at the expense of more false positives; any increase in the burden to bring suit will reduce false positives at the expense of more false negatives.

Anyway, Ted continues, shareholders of NJ corporations can decide to invest in firms incorporated elsewhere if they think NJ law is too lenient on directors, aptly citing my and O’Hara’s The Law Market.

Of course Frankel might argue that the business judgment rule that the court used to decide the case is ubiquitous, leaving plaintiffs with little choice. Indeed, the only significant dissent is Nevada which is, if anything, even easier on directors than NJ.   Frankel might also argue that this indicates state corporation law is rigged for managers and that we would do better under federal law.  Perhaps what we need is a super Dodd-Frank/SOX on steroids that preempts state law and exposes managers to suits like the one NJ dismissed.

I would respond that the universal acceptance of the business judgment rule represents the market’s rejection of Frankel’s position.  If Frankel wants to complain that the market for corporate law is imperfect,  she would need to persuade me that shareholders are better off in the clutches of Congress.

In upholding the board’s power, and confirming what most astute observers knew the law likely was, despite very strong facts the other way (no “coercion” or other obvious reason why the pill was needed), the court preserved the stability of Delaware law. Hard facts don’t have to make bad law.  Although the Air Products bid was $70, the shares had been trading around $63.  They fell from $63.73 to $61 after the decision, suggesting clarification of a small residual amount of uncertainty.

The decision illustrates one reason why Subramanian, et al, were wrong to suggest the Delaware anti-takeover statute is preempted by the Williams Act.  If the statute, why not the pill — and other aspects of Delaware law that block takeovers?  Here’s my response along these lines to the Subramanian et al argument.

Of course Congress theoretically could pass a law that explicitly preempts poison pills.  Dodd-Frank’s intrusion into corporate governance makes this plausible.  But such additional federal interference would not be wise.

In Jones v. Harris Associates, the Supreme Court interpreted investment advisers’ fiduciary duty regarding compensation for services under Section 36(b) of the Investment Company Act of 1940. The Court endorsed an open-ended Second Circuit standard over a more determinate Seventh Circuit test calling only for full disclosure and no “tricks.”  This paper shows that Congress created and must solve the fundamental problem the Court faced in Jones. At one level the problem stems from the existence of an investment adviser fiduciary duty as to compensation and the corporate structure from which it springs. At a deeper level lies the more basic problem of federal interference in firm governance, which lacks state corporate law’s safety valve of interstate competition and experimentation.  This discussion is appropriate in light of the increasing federal role evident in the enactment of broad new financial regulation.

Here’s an earlier pre-decision post laying out some of the issues, which I concluded by noting:

I suspect that in this day and age the Supreme Court will side with Posner. Such a decision would be a symptom and signal of our sharp turn toward paternalism in everything from complex finance to corporate governance to the simplest products.

Although the Court’s result was consistent with Posner’s position, as noted here it left the deeper issues to Congress. These include questions not only about markets, but also about the appropriate federal role in structuring investment vehicles. My article suggests that, whatever the flaws in markets and state regulation, federal regulation may be worse.

Regulation of beer, wine and liquor in the post-Prohibition world largely consists of a patchwork of state laws that benefit wholesalers.  Many of these restrictions you are familiar with.  Others, however, are less well known.  For example, most states mandate a “three-tier” system in which the brewer/winery must sell to the wholesaler, who in turn, must sell to the retailer.  In other words, the laws restrict vertical integration between levels.  Aside from laws on direct shipment impacted by the Commerce Clause challenge in Granholm, and on top of the three tier system, most states also adopt laws that restrict the contractual relationships between alcohol producers and wholesalers.  For example, most states have “just cause” laws that limit (sometimes to the point of effectively prohibiting) the producers ability to terminate the wholesaler.  States also sometimes mandate that the wholesaler be given an exclusive territory by the producer, or ban a producers ability to enter into exclusive dealing arrangements with wholesalers.  Unsurprisingly, many of these state laws harm consumers in the form of allowing higher prices and reduced output.  Therein lies the rub.  In Granholm, the Court explained that the “aim of the Twenty-first Amendment was to allow States to maintain an effective and uniform system for controlling liquor by regulating its transportation, importation, and use.”  Anticompetitive state regulation which raises barriers to entry, reduces competition, and increases price also reduces output.  And while reduced alcohol consumption is a consumer welfare loss in the conventional antitrust sense, it may count as a benefit under a 21st Amendment analysis.

While wine lovers have had all the fun in terms of litigating state alcohol regulation, and most (but not all) of the debate has been about direct shipment laws, state laws greatly impact that distribution, marketing, and pricing of beer, wine and liquor.  There is also a lot more at stake than just direct shipment laws.  The state laws restricting contractual relationships between producers and wholesalers have significant competitive effects and are prevalent across nearly all states and for all forms of alcoholic beverages.  In other words, beer, wine and liquor lovers can unite in their disdain for these laws or federal legislation that places them out of the reach of any challenge.  So, this time, as Mercatus’ Jerry Ellig notes, its not just the wine wholesalers that are pushing the law:

Word on the street is that the biggest pushers of this legislation are the beer wholesalers. Since most of this litigation has involved wine, what’s going on here?

The real goal of this legislation is not harrassing wineries that want to ship a few bottles to out-of-state customers. The real goal is to preserve anti-competitive state laws that force brewers, wine makers, and distillers to market most of their product through beer, wine, and spirits wholesalers, instead of marketing directly to retailers and restaurants. The proposed legislation would effectively insulate these state laws from challenge under the Commerce Clause, federal antitrust laws, or any other federal laws that might give alcohol producers and consumers some leverage to break the wholesalers’ lock on the market.

Call it states’ rights kool-aid with a chaser of economic protectionism.  A strange brew indeed.

That takes us to the CARE Act.  As Katherine Mangu-Ward of Reason explains:

The legislation would privilege the 21st Amendment, which repealed Prohibition and left regulation of alcohol to the states, over the Constitution’s Commerce Clause, which prohibits states from favoring in-state entities in trade. The bill would exempt state booze laws from federal limits—stuff like anti-trust rules and Food and Drug Administration requirements. It would allow states to replace uniform federal standards with their own competing standards for labeling and formulation, making it harder for out-of-state producers to comply with state laws. It would also take away recourse to the courts for sellers or buyers harmed by the new legislation. At present, the burden of proof is on the state to show that it needs to discriminate against outside competitors for a legitimate purpose, such as the health or safety of its citizens. This would be reversed, requiring anyone who wanted to challenge a state alcohol law to first prove that the law has no legitimate purpose beyond protectionism, a nearly impossible task.

A look at the text of the CARE Act confirms Mangu-Ward’s analysis.  Here’s Section 3(b)(c) of HR 5034, where the damage is done:

The following shall apply in any legal action challenging, under the Commerce Clause or an Act of Congress, a State or territory law regarding the regulation of alcoholic beverages:

(1) The State or territorial law shall be accorded a strong presumption of validity.

(2) The party challenging the State or territorial law shall in all phases of any such legal action bear the burden of proving its invalidity by clear and convincing evidence.

(3) Notwithstanding that the State or territorial law may burden interstate commerce or maybe inconsistent with an Act of the Congress, the State law shall be upheld unless the party challenging the State or territorial law establishes by clear and convincing evidence that the law has no effect on the promotion of temperance, the establishment or maintenance of orderly alcoholic beverage markets, the collection of alcoholic beverage taxes, the structure of the state alcoholic beverage distribution system, or the restriction of access to alcoholic beverages by those under the legal drinking age.’’

That’s quite some burden!  Let’s talk about the Sherman Act and preemption here.  Lets say that a party wanted to challenge anticompetitive state regulation on the grounds that it is preempted by the Sherman Act.  Notice that the party challenging the state law would not only have to satisfy the preemption standard, but also demonstrate by clear and convincing evidence that the state law has “no effect” on the promotion of temperance, the orderliness of alcoholic beverage markets, the collection of taxes, market structure, or underage drinking.  If the state law impacts any of those, it survives.  Considering that the purpose of these laws is to raise prices and reduce output, such laws are likely to have at least some minimal effect on temperance, depending on how that term is defined.  If temperance is construed as merely meaning “any” reduction in consumption, than the law of demand implies that anticompetitive state regulations always promote temperance.  That’s absurd.  More likely, temperance must correlate with the social ills associated with drinking, e.g. drunk-driving, health problems, accidents and fatalities. Under that interpretation, it is unclear whether the laws will promote temperance because they will raise the price both to marginal drinkers with higher demand elasticities and infra-marginal drinkers with more inelastic demand.  But no effects?  Taken literally, the law of demand implies that unless the drinkers who create those social harms have perfectly inelastic demand, there will be “some” effect and thus the state law shall be upheld.

But notice how the next clause of 5034(3)(b)(c) requires the challenger to demonstrate that the law has no impact on the maintenance of “orderly alcoholic beverage markets.”  What does this mean anyway?  Antitrust lawyers will recognize “maintenance of orderly” markets as a codeword for keeping prices stable and high.  Well, if that is what orderly means here, then this clause swallows the rest of the rule and all challenges, whether based on the antitrust laws or the Commerce Clause, are doomed to failure.

Realistically, that seems to be what is at stake here: an antitrust exemption for beer and wine wholesalers.  I’ve written before about the weak case for antitrust exemptions.  But the case cannot possibly be weaker than when the cartel is backed by the coercive power of state law.  Unfortunately, the proposal seems to be getting some traction, including a Judiciary Committee Hearing on the subject earlier this year (Darren Bush’s (Houston) testimony is recommended reading).  An exemption for beer and wine wholesalers simply doesn’t make sense on competition policy grounds unless it can be shown that the social harms associated with reduced consumer welfare caused by higher prices and reduced output is more than offset by gains from “promoting temperance.”

A serious cost-benefit analysis of state regulations of the alcoholic beverage industry raises empirical questions concerning these tradeoffs.  We know that consumers are harmed from higher prices and reduced output.  But it is possible that those effects are dominated by benefits associated with reduced consumption leading to reduced social harms.  The CARE Act legislates away an rigorous attempt to deal with these tradeoffs, and bets consumer dollars in the process, by assuming that the reduction in alcohol consumption is “always worth it.”   But there is little empirical evidence that many of these state restrictions are associated with social benefits.

Later this week, I’ll follow up with a post discussing some of my own recent empirical research on this topic.

Now Annette Nazareth summarizes (HT Bainbridge) the provisions in the bill that passed the Senate and awaits reconciliation. She notes that the bill “would federalize significant governance and executive compensation matters that have historically been a matter of state law.” Alas, the Senate never voted on an amendment proposed by Delaware’s Carper that would have eliminated (D-Del) that would have eliminated the majority voting provision and a provision for proxy access.

Although none of the provisions Nazareth discusses is individually earth-shaking, they cumulatively touch many major aspects of corporate governance formerly left to contract and state law.  This bill thus clearly adds to the framework for federal takeover of internal governance that SOX established. The overall effect is that it will be increasingly difficult to demark an area left exclusively for state law. This leaves little “firebreak” to protect against judicial incursions in the spaces not yet covered by explicit federal provisions.  This could ultimately profoundly affect the relationship between federal and state law regarding business associations. 

A generation ago the Supreme Court could say that “no principle of corporation law and practice is more firmly established than a State’s authority to regulate domestic corporations, including the authority to define the voting rights of shareholders.” CTS Corp. v. Dynamics Corp. of America, 481 U.S. 69, 89 (1987). 

Erin O’Hara and I have argued that this separation between federal and state spheres does and should affect the scope of implied preemption of state law by federal statutes.  Thus, when the Court held that state securities actions were preempted by the Securities Litigation Uniform Standards Act, it emphasized “[t]he magnitude of the federal interest in protecting the integrity and efficient operation of the market for nationally traded securities.” Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Dabit, 547 U.S. 71, 78 (2006). See also my article on Dabit. However, we noted that “[m]any federal ‘securities’ laws reach deep into the kind of internal governance issues covered by the [internal affairs doctrine].” Thus, corporate internal affairs are only “relatively safe from federal preemption” and internal affairs is not “a constitutional boundary, as shown by the continuing forward march of federal corporation law.”

Under the Dodd bill, the forward march picks up the pace.  

Yet from a policy standpoint the march is very much backward. In my April post I observed that “[a]s financial markets have become far deeper and more competitive since the 30s, it makes little sense for regulators to actually trust them less.” Thus, the Senate has ignored not only the lessons of SOX but the developments in corporate governance and markets that make its governance provisions less necessary than ever.

On one side of this debate, Commissioners Rosch and Leibowitz have extolled the virtues of Section 5 as  a zone free of collateral consequences.   Indeed, Chairman Leibowitz and Commissioner Rosch have gone so far as to assert that the logic underlying the Supreme Court’s jurisprudence recognizing the social costs of antitrust error should only apply to private plaintiffs, but not the enforcement agencies because the latter know anticompetitive conduct when they see it while generalist judges and juries do not and thus are more prone to costly errors.

As a preliminary matter, I should note that my view is that the Rosch/ Leibowitz claim that antitrust law has been narrowed exclusively because of concerns about private plaintiffs is dramatically overstated at best, and at worst, blatantly inconsistent with the Supreme Court’s jurisprudence which has been remarkably consistent in focusing on the inherent difficulties associated with identifying anticompetitive conduct when discussing error costs and the role they play in setting antitrust rules.  It is certainly true that courts have expressed concerns about abuse of the antitrust laws by private plaintiffs, but it is impossible to read the Supreme Court’s antitrust jurisprudence (see, e.g. Trinko, Nynex, Credit Suisse, Brooke Group, Linkline) without taking away a much the Court’s much more general fear that the social costs of false positives — stemming both from the burdens of antitrust discovery and chilling of efficient, pro-competitive conduct — warrants a reduction in the scope of the antitrust laws.

Holding that point aside for a moment, the primary argument supporting the controversial expansion of Section 5 has been that it does not have collateral consequences because only Section 2 create follow-on opportunities for private plaintiffs.  Section 5, Chairman Leibowitz and Commissioner Rosch tell us, does not present such problems.  The social costs associated with Section 5 follow-ons do not exist. Here is a recent version of that assertion from Commissioner Rosch:

The problem here, however, is that these Supreme Court decisions adversely impact all cases in which public antitrust enforcers proceed under the same statute as private plaintiffs and state enforcers, including, most significantly Sections 1 and 2 of the Sherman Act. The Commission can avoid implicating both of these concerns if it proceeds under Section 5: first, only the Commission (as divorced from private plaintiffs, for example), can proceed under Section 5 and, second, if the Commission
proceeds under its Part 3 administrative process in a Section 5 proceeding, there is no role for the district courts or federal juries to play.

Here’s another version of the case for Section 5 resting on an empirical claim about the lack of collateral consequences:

This is an especially important consideration when federal court private treble damage litigation involving the same conduct is pending or threatened. But is it important whenever there is a reasonable prospect that such a private claim will be filed. A plaintiff cannot rely on favorable Section 5 case law in a federal treble damage action. Neither can a federal district court rely on such a decision because the FTC alone can avail itself of Section 5 at the federal level. Conversely, the spillover effects on federal law enforcement of Supreme Court substantive law jurisprudence that is the product of concern about such treble damage actions can be reduced if the Commission uses Section 5, instead of traditional antitrust law that is equally applicable to private and public plaintiffs.

Whether an expanded Section 5 would lead to collateral consequences is fundamentally an empirical question.  There are two potential sources of collateral consequences that have been discussed.  The first is private actions using FTC Act settlements or judgments to create claims under state consumer protection acts (CPAs), which are often interpreted in light of the FTC Act and allow for both attorneys’ fees and multiple damages. The second is the possibility of follow-ons in which private plaintiffs rely on the Section 5 settlement or judgment in federal court under a traditional antitrust statute such as Section 1 or 2.

Thus far, the debate has focused on the first source: state CPAs.  Commissioner Kovacic first responded to the aggressive use of Section 5 in his dissent in the N-Data settlement, responding to the majority assertion that Section 5 provides a free lunch:

The Commission overlooks how the proposed settlement could affect the application of state statutes that are modeled on the FTC Act and prohibit unfair methods of competition (“UMC”) or unfair acts or practices (“UAP”). The federal and state UMC and UAP systems do not operate in watertight compartments. As commentators have documented, the federal and state regimes are interdependent. [Citations omitted].  By statute or judicial decision, courts in many states interpret the state UMC and UDP laws in light of FTC decisions, including orders. As a consequence, such states might incorporate the theories of liability in the settlement and order proposed here into their own UMC or UAP jurisprudence. A number of states that employ this incorporation principle have authorized private parties to enforce their UMC and UAP statutes in suits that permit the court to impose treble damages for infringements.

As a matter of theory, because treble damage remedies are not generally available for Section 5 violations, overdeterrence is likely to be less of a problem under Section 5 than Section 2. But make no mistake — the difference is one of degree rather than of kind.  And as a practical matter, the difference virtually disappears when private remedies are available under Little FTC Acts, including those that are construed in harmony with Section 5 and allow for multiple damages and attorneys’ fees.

In response to this argument, the Commissioners favoring expansion of Section 5 have played the unicorn card.  The claim is, quite simply, that such state level follow-ons don’t happen.  Unfortunately, this side of the debate has been a data-free zone thus far.  Well, almost data free.  Much has been made about the fact that the N-Data settlement itself did not give rise to private causes of action under any “Little FTC Acts.”  For example, Chairman Leibowitz has pointed to the fact that no plaintiff in N-Data filed under a state consumer protection act as evidence that “Section 5 violators do not find themselves subject to private antitrust actions under federal law— and probably under state baby FTC acts as well—certainly not for treble damages.”  Well, state consumer protection acts do indeed exist.  And there is, as the Searle Report on Private Litigation under CPAs notes, quite a bit of litigation under them.  A systematic empirical evaluation of state CPA litigation to determine how frequent Section 5 follow-on litigation occurs seems like a superior alternative to the current debate.

As a tangent, I find the theoretical underpinnings of the “it doesn’t happen” response both unpersuasive and a little bit odd without systematic data supporting it.  The argument seems to be that the private plaintiffs bar is insufficiently creative, resourceful, or aggressive enough to make use a perfectly operational statute that allows free-riding on the Commission’s efforts, and access to attorneys’ fees and multiple damages.  Has anybody seen a modern public policy debate in which it has been asserted that the private plaintiffs bar is asleep behind the wheel?  Now I’m the first to say that measuring the extent and magnitude of any collateral effects of Section 5 is an important empirical question that really ought to be addressed in detail with a serious study.  But pointing to a particular case (N-Data) where a particular plaintiff chose not to make use of the statute for any number of reasons as evidence that it “doesn’t happen” despite obvious incentives for private plaintiffs to use state acts doesn’t quite do the trick.  The appropriate default presumption, I’d imagine, should be that the plaintiffs bar makes use of these statutes where it is profitable to do so.

Of course — state courts are not the only source of follow-ons.  One might believe that a Commission action under Section 5 would encourage private plaintiffs to proceed under Section 2 (or some other statute, e.g. Robinson-Patman) as well.  We are told that such a possibility is pure theory and a Chicago School figment of the imagination.  Well … not so fast.

Readers might recall that the Commission recently announced a settlement with Transitions Optical in a Section 5 case involving exclusive dealing contracts.  A private class action suit alleging a violation of Section 2 of the Sherman Act has been filed in the Western District of Washington based on the Commission Section 5 settlement.

The class action suit comes a month after the Federal Trade Commission (FTC) reached a settlement with Transitions Optical that bars Transitions Optical from using allegedly anticompetitive practices to maintain its monopoly and increase prices on photochromic lenses. The settlement was accompanied by a “consent agreement” between the FTC and Transitions Optical that includes restrictions on exclusive or preferred customer relationships. Transitions Optical has denied any wrongdoing in the matter. … A Transitions Optical spokesperson said, “It’s not unusual for lawsuits like this to be attempted after consent agreements like ours are filed with the FTC. We remain confident about our company and our business practices and feel that we have always done what is right for the success of all our customers and partners. The continued support we receive from our customers and partners shows, more than anything, that we are working in the best interest of the optical industry overall.”

Similarly, a class action Section 1 complaint has been filed against Guitar Center and the National Association of Music Merchants in the Northern District of Illinois.  In that complaint, which relies heavily on the FTC Section 5 complaint, consent order and press releases throughout, the plaintiff alleges that defendants’ minimum advertised pricing (MAP) policies resulted in collusion and higher prices.

The fact that Section 5 judgments do not automatically result in liability under Sherman Act Section 1 or 2 in federal court makes these follow-ons different in at least some important ways from the state CPAs, where at least in principle, liability is automatic.  However, the debate over whether Section 5 consents are free of collateral consequences doesn’t turn on automatic liability.  Rather, the debate is over whether and to what extent Section 5 consents and judgments generate collateral consequences in the form of follow-ons that can lead to the same treble damages actions that generate the concerns about socially costly false positives in the Section 2 setting.  All parties apparently agree that the the Supreme Court’s jurisprudence reacting to the problem of false positives is a sound and rational approach.   The two federal court cases revealed by a few Google searches, combined with the possibility of state CPA actions which have the advantage of multiplied damages along with automatic liability, suggests that we are not talking about unicorns here. The ratio of theory to evidence in the policy discussion of error costs and Section 5 is greater than optimal.  But the claim that the Commission’s expanded vision of Section 5 is free of those concerns should be subjected to more rigorous empirical testing before accepted as a sound basis for competition policy.

The courts have pared back plaintiffs’ rights in antitrust cases. They’re concerned about what they believe to be the toxic combination of class actions, treble damages and a very aggressive plaintiffs’ bar. The problem for us as an agency is we come under those restrictions, [too]. So how do we do what we’re supposed to do, which is stopping anticompetitive behavior? One tool in our arsenal is using what’s known as our Section 5 authority to stop unfair methods of competition.

Leibowitz further justifies his approach to Section 5 with an appeal to what he claims to be an important intrinsic limit of Section 5:

The other advantage of this authority is, because it’s not an antitrust statute, it’s going to limit follow-on, private treble-damages law suits. I think in the end, if we use this statute effectively to stop anticompetitive behavior, the business community is going to end up supporting it very, very strongly. Because what they’re most concerned about is follow-on, private, treble-damages litigation. They’re not so much concerned about cease-and-desist [orders], which is the kind of thing we’re often looking at when we use our Section 5 authority. I don’t think big business should be worried. I think they should embrace this trend.

Yes, I’m sure business will eagerly embrace the FTC’s use of this statute, particularly as the agency defends it precisely on the ground that its use is relatively unconstrained by courts and their pesky rule of law.

Leibowitz has been making these claims for some time (see, e.g., these remarks from October 2008 and the N-Data Statement).

But admittedly, if it were true that the FTC’s use of Section 5 did not lead inexorably to costly follow-on litigation, and if it were not the case that the statute were a recipe for unprincipled, uneconomic antitrust enforcement, no doubt there would be some support for it.  But unfortunately for Leibowitz, the claim is NOT true–it is not the case that Section 5 removes the specter of costly private litigation from the equation.

The reality is that many states have “Baby FTC Acts,” modeled on the federal FTC Act and taking enforcement cues–by law–from FTC interpretation of the Act.  And these statutes do provide for private rights of action and treble damages.  So although it is technically true that there is no private right of action under the federal FTC Act, this hardly shields antitrust defendants from follow-on liability.  And even if such actions have been rare up until now (as Leibowitz claims in the remarks linked above), that may well change as the FTC’s precedent-setting enforcement decisions shift toward using the statute as an antitrust enforcement tool and as the Act is used more and more for otherwise-unwinnable Sherman Act cases.

This point isn’t new, and Commissioner Kovacic made this same point in his dissent from the N-Data settlement:

The Commission overlooks how the proposed settlement could affect the application of state statutes that are modeled on the FTC Act and prohibit unfair methods of competition (“UMC”) or unfair acts or practices (“UAP”). The federal and state UMC and UAP systems do not operate in watertight compartments. As commentators have documented, the federal and state regimes are interdependent. [Citations omitted].  By statute or judicial decision, courts in many states interpret the state UMC and UDP laws in light of FTC decisions, including orders. As a consequence, such states might incorporate the theories of liability in the settlement and order proposed here into their own UMC or UAP jurisprudence. A number of states that employ this incorporation principle have authorized private parties to enforce their UMC and UAP statutes in suits that permit the court to impose treble damages for infringements.

If the Commission desires to deny the reasoning of its approach to private treble damage litigants, the proposed settlement does not necessarily do so. If the Commission’s assumption of no spillover effects is important to its decision, a rethink of the proposed settlement and order seems unavoidable.

As far as I can tell, however, Leibowitz and other defenders of this rationale for expanded Section 5 enforcement have not addressed this point, and they continue to rely, disingenuously, in my opinion, on claims that Section 5 enforcement will not lead to follow-on, private actions.

At the same time, as I pointed out here, Leibowitz’ continued claim that courts have reined in Sherman Act jurisprudence only out of concern with the incentives and procedures of private enforcement, and not out of a concern with a more substantive balancing of error costs–errors from which the FTC is not, unfortunately immune–seems ridiculous to me.  To be sure (as I said before), the procedural background matters as do the incentives to bring cases that may prove to be inefficient.

But take, for example, Twombly, mentioned by Leibowitz as one of the cases that has recently reined in Sherman Act enforcement in order to constrain over-zealous private enforcement (and thus not in a way that should apply to government enforcement).  Yes, of course, Twombly was concerned with the private incentives for bringing antitrust strike suits and the costs of such suits.  (And I note in passing that, while the specific monetary incentive at issue in the case might not apply to the government, the government, too, certainly has incentives to bring cases that may be weak–I hardly think the analysis is completely inapposite.  Meanwhile the costs of protracted litigation are just as high if the plaintiff is the government as if it is a private party.)

But the over-zealousness of private plaintiffs is not all it was about, as the Court made clear:

The inadequacy of showing parallel conduct or interdependence, without more, mirrors the ambiguity of the behavior: consistent with conspiracy, but just as much in line with a wide swath of rational and competitive business strategy unilaterally prompted by common perceptions of the market.  Accordingly, we have previously hedged against false inferences from identical behavior at a number of points in the trial sequence.

* * *

Hence, when allegations of parallel conduct are set out in order to make a §1 claim, they must be placed in a context that raises a suggestion of a preceding agreement, not merely parallel conduct that could just as well be independent action. [Citations omitted].

The Court was appropriately concerned with the ability of decision-makers to separate pro-competitive from anticompetitive conduct.  Even when the FTC brings cases, it and the court deciding the case must make these determinations.  And, while the FTC may bring fewer strike suits, it isn’t limited to challenging conduct that is simple to identify as anticompetitive.  Quite the opposite, in fact–the government has incentives to develop and bring suits proposing novel theories of anticompeitive conduct and of enforcement (as it is doing in the Intel case, for example).

I recognize that Leibowitz may believe that he is not susceptible to mistakes of this sort, or that (as Dan Crane might say), the FTC has a comparative institutional advantage over courts in making these sorts of determinations.  I disagree, but if that is the claim then Leibowitz should make it explicitly rather than suggesting that current Sherman Act jurisprudence is all about treble damages and strike suits.  I’m quite certain, however, that an explicit claim by the FTC that it never gets it wrong and thus shouldn’t be constrained by meddling courts wouldn’t be viewed very favorably by the business community.

Here’s the abstract:

This essay reviews Michael Carrier’s analysis of antitrust and standard setting in his new book: Innovation for the 21st Century: Harnessing the Power of Intellectual Property and Antitrust Law. While Innovation for the 21st Century offers a balanced and informative summary on patent holdup, we find that Carrier’s treatment of antitrust and standard setting avoids too many of the critical policy questions. One critical and emerging issue in this area, and one Professor Carrier largely ignores, is the use of Section 5 of the FTC Act to govern the standard setting process, as in In re N-Data. We explore and highlight some of the critical legal and economic issues associated the use of Section 5 in the patent holdup context, the standard courts should apply to this conduct under Section 2 of the Sherman Act, and the fundamental issue of whether innovation and economic growth would be better served by relying on contract and patent law rather than antitrust. We conclude that it is highly unlikely that optimal regulation of standard setting activity includes the creation of perpetual contractual commitments backed by the threat of antitrust and state consumer protection remedies, without rigorous economic proof of substantial consumer injury that cannot be reasonably avoided. In our view, the current state of affairs described herein presents a critical threat to standard setting activity and innovation.

The essay can be downloaded here.

Were this a full book review and not merely a blog post, I would spend more time identifying the many points in the book that I agree with. But it is not. Instead, I will narrow my focus to Professor Carrier’s approach to standard setting activities, and in particular, patent holdup. Chapter 14 is largely devoted to summarizing the state of affairs in antitrust and standard setting. The summary (pages 323-342) is balanced, well-written and recommended reading for anyone interested in getting up to speed on the current policy issues. After summarizing Professor Carrier’s proposal for antitrust analysis of patent holdup (and other business conduct in the standard setting process), I’ll turn to highlighting a few areas where I found myself either disagreeing with his analysis or hoping for a more complete treatment.
In my own view, the two most pressing policy issues with respect to patent holdup are:

1. What is the appropriate role of antitrust in governing patent holdup?

2. If antitrust rules should govern patent holdup, which statute(s) and what type of analysis should apply? In particular, what is the appropriate scope of Section 2 of the Sherman Act and Section 5 of the FTC Act?

While Professor Carrier’s treatment of patent holdup usefully summarizes the debate, and also recommends a policy proposal that I largely agree with, I was left hoping for a bit more in this section of the book in terms of moving the ball forward on these important questions.

Let’s begin with the policy proposal itself. Professor Carrier argues that “given SSOs significant pro-competitive justifications, courts and the antitrust agencies should consider their activity under the Rule of Reason.” Carrier carves out standard setting organization (SSO) members’ joint decisions to fix prices on the final goods sold to consumers as the only conduct deserving of per se treatment. So far I’m on board. It makes economic and legal sense to treat both standard setting activities (with the exception of cartel behavior) and IP rules of SSOs as generally procompetitive and thus falling under the rule of reason. Carrier identifies three potential areas of liability concern under the rule of reason: patent holdup (he cites Dell and Unocal as examples), boycotts, and situations in which SSOs exert buyer power to reduce prices with the effect of reducing the incentive to innovate. Carrier writes that “absent these situations, SSO activity should be upheld under the rule of reason.”

There is much I agree with here. In fact, I find myself in agreement with Professor Carrier about most of what he writes about the limited utility of per se analysis in the standard setting arena. But I will focus on some areas where I suspect that we disagree, though I’m left unsure based solely on what is in the book. Carrier identifies patent holdup involving deception as a cause for concern under a rule of reason analysis. But the treatment is cursory. Carrier writes that “such activity could demonstrate attempted monopolization under Section 2 of the Sherman Act” and notes that a plaintiff making such a claim must demonstrate, amongst other requirements, that “the deception result[ed] in a standard’s adoption or higher royalties.” (page 342).

It is helpful for my purposes to bifurcate the world of patent holdup theories into those involving deception (the stylized facts in Rambus or the allegations in Broadcom v. Qualcomm) from those that do not and merely involve the ex post modification and/or breach of contractual commitments made in good faith in the standard setting process (FTC v. N-Data). Again, with respect to each of these patent holdup theories, there are at least two critically important policy issues:

1. What is the appropriate role of antitrust in governing patent holdup?

2. If antitrust rules should govern patent holdup, which statute(s) and what type of analysis should apply? In particular, what is the appropriate scope of Section 2 of the Sherman Act and Section 5 of the FTC Act?

With respect to the first policy question, Carrier appears to presume that antitrust rules should apply to unilateral conduct in the form of patent holdup involving both deception and breach theories. I may be wrong about the breach theories. While Carrier discusses N-Data briefly, his policy proposal singles out examples such as Dell and Unocal, which involved deception. I was left wanting a more clear exposition of the details of the policy proposal in this section. More fundamentally, the relative merits of state contract law and the patent doctrine of equitable estoppels in the SSO setting as alternatives to antitrust liability are an important topic. Of course, this issue is one of special concern for me since Kobayashi and Wright (Federalism, Substantive Preemption, and Limits on Antitrust) have argued that antitrust rules layered on top of these alternative (and we argue superior) regulatory institutions threaten to chill participation in the SSO process and reduce welfare. But Kobayashi and Wright are not alone in questioning the utility of antitrust liability layered on top of these alternative bodies of state and federal law. For example, Froeb and Ganglomoir present a model in which “the threat of antitrust liability on top of simple contracts shifts bargaining rents from creators to users of intellectual property in an inefficient way.” Other contributors to this literature questioning the role of antitrust liability in “breach” style patent holdup cases such as N-Data include Anne Layne-Farrar. I will not take on the task in this blog post of repeating the various arguments making the case against antitrust liability here. But I believe that Carrier’s standard setting chapter and policy proposals would benefit from addressing them.

Second, assuming that antitrust rules should apply to patent holdup (both deception and breach variants), what should the analysis look like? With respect to the Section 2 analysis in claims involving deception, Professor Carrier appears to endorse the proposition that a demonstration of either actual exclusion (e.g., the deception is the but-for cause of the adoption of the technology) or higher royalties would be sufficient to support such a claim. I’m not sure why the latter is or should be sufficient? As I’ve argued elsewhere, the Supreme Court’s decision in NYNEX applies in the patent holdup setting when (1) the patent holder has market power prior to the deception and (2) the deceptive conduct results in higher royalties but not exclusion of rival technologies. When those conditions are satisfied, NYNEX’s holding (which is consistent with much of the Supreme Court’s general jurisprudence about the monopolist’s freedom to optimal pricing, e.g., Trinko, Linkline) that deceptive or fraudulent conduct that merely results in higher prices but not exclusion cannot be the basis of a Section 2 claim. Along those lines, I’ve argued that the D.C. Circuit’s Rambus decision is best interpreted as calling the Commission to task for failure to meet its burden of demonstrating that the first of these conditions did not apply. In any event, in reading Carrier’s treatment of patent holdup issues I’m left with several questions. For instance, I’m left wondering whether he believes that Section 2 should apply to both the deception and breach variants of patent holdup? If it applies to both, what is the appropriate scope of NYNEX? For example, do plaintiffs in patent holdup claims under Section 2 have the burden of demonstrating that the patent holder did not have monopoly power prior to the deceptive conduct? If not, on what grounds is NYNEX distinguishable? Is it because it was not an SSO case? What is the appropriate rule of reason analysis in a case involving deception in the standard setting process? What about cases like N-Data where the plaintiff does not allege any “bad conduct” at the time the technology is selected by the standard but rather some renegotiation of contract terms at a later time?

Third, no discussion of patent holdup would be complete without a discussion of whether and how Section 5 of the FTC Act should apply to patent holdup theories. Here again, while Carrier discusses N-Data briefly, this question does not receive attention. So the blog symposium seems like a great place to ask questions like the following: Should Section 5 of the FTC Act apply to both the deception-based and the “pure breach” variants of patent holdup? These are some of the most pressing issues relating to antitrust analysis of standard setting. Recently, Chairman Leibowitz singled out N-Data as a paradigmatic example of the appropriate application of Section 5:

One category of potential cases [to which to apply Section 5] involves standard-setting. N-Data, our consent from last spring, is a useful example. Reasonable people can disagree over whether N-Data violated the Sherman Act because it was never clear whether N-Data’s alleged bad conduct actually caused its monopoly power. However, it was clear to the majority of the Commission that reneging on a commitment was not acceptable business behavior and that—at least in this context—it would harm American consumers. It does not require a complex analysis to see that such behavior could seriously undermine standard-setting, which is generally procompetitive, and dangerously limit the benefits that consumers now get from the wide adoption of industry standards for new technologies.

Tales from the Crypt” Episode ’08 and ’09: The Return of Section 5 (“Unfair Methods of Competition in Commerce are Hereby Declared Unlawful”).
Similarly, Commissioners Leibowitz, Rosch, and Harbour noted in the N-Data majority statement that “there is little doubt that N-Data’s conduct constitutes an unfair method of competition,” describing the renegotiation of the ex ante contractual commitment to license at $1,000 to a RAND commitment as “oppressive” and an act that threatens to “stall [the standard setting process] to the detriment of all consumers.”

I wonder whether Professor Carrier thinks the majority in N-Data was correct? And if so, on what basis?  Or are breach variant holdup claims more appropriately governed under Section 2? If the answer to either of those questions is yes, I’d like to know whether and on what basis the application of these mandatory antitrust rules is superior to contract law, which contains doctrine designed to identify and distinguish good faith modifications and renegotiations from attempts at ex post opportunism.

I should note that I do not consider it a criticism of the book that these details are largely left out of the book. The task of organizing a coherent and intellectually provocative book that moves between copyright, patent, and antitrust is monumental and comes with its own special set of breadth and depth tradeoffs. However, I ultimately found the attention to legal, economic, and policy details in the SSO section less satisfying than the treatment of other equally complex issues throughout the book. While I was left disappointed that these details were not there, I admit to being very curious after reading Professor Carrier’s views on innovation and antitrust more generally as to how he will manage the thorny details in the patent holdup context.