By: Dorit Rubinstein Reiss


[N]ot all capture is bad.  It surely is bad for regulators to believe that ‘what’s good for General Motors is good for America’; but it is also undesirable for regulators to believe that ‘what’s bad for General Motors is of no consequence to America’.[1]

Observers of the administrative state warn against capture of administrative agencies and lament its disastrous effects.  This Article suggests that the term “capture,” applied to a close relationship between industry and regulator, is not useful; by stigmatizing that relationship—judging the relationship as problematic from the start—the stigmatization hides the relationships potential benefits.  The literature on capture highlights its negative results: lax enforcement of regulation, weak regulations, and illicit benefits going to industry.  This picture, however, is incomplete and in substantial tension with another current strand of literature which encourages collaboration between industry and regulator.  The collaboration literature draws on the fact that industry input into the regulatory process has important benefits for the regulatory state.  Industry usually has information no one else has and has more incentive to give that information to a friendly regulator.  Furthermore, working with industry can substantially improve the impact of regulation; voluntary compliance is cheaper and can be more effective than enforced compliance, and industry can help regulators minimize negative unintended consequences.  This Article suggests that instead of engaging in name calling, we should focus on identifying when a close industry-regulator relationship will work in the public interest and when it is likely to undermine it.  That is an empirical question.

For decades, scholars have discussed capture of administrative agencies—mainly, though not exclusively, by industry[2]—strictly in terms of the negative consequences.[3]  If “accountability” is often seen as the “hurrah word” of the administrative state,[4] capture can be seen as the “boo-word”; historically, up until quite recently, few have had anything good to say about it.[5]  The term capture itself is a discussion ender; if an agency is said to be “captured,” the regulatory results are presumed to be bad.  Like most negative labels, this presumption tends to obstruct efforts to arrive at the kind of clear understanding that leads to good policy prescriptions.

Capture refers to an extremely close relationship between regulators and industry.[6]  Some believe such a relationship is inherently dangerous and negative.  Reports prepared by Ralph Nader’s associates, the “Nader’s Raiders,”[7] reflect that spirit.  For example, the Nader report on the Food and Drug Administration (“FDA”) saw the FDA as a puppet in the hands of industry and claimed that “when the law allows administrative discretion, industry, not the consumer, benefits.”[8]  It evaluated the FDA’s supervision of firms and criticized its presumption that almost all manufacturers can be trusted to place the public interest before profit.[9]  The report strongly criticized the FDA’s decisions to attempt to educate firms about the need for compliance rather than punish them for noncompliance.[10]

In the same vein, the group’s study of the Federal Aviation Administration (“FAA”) accused the regulators of being so closely related to the industry that they promoted the industry’s interests over consumers’ safety.[11]  The problem with this approach is that it is one sided.  Yes, a cozy relationship between regulator and industry creates risks and may (though not inevitably) lead to terrible outcomes.  But an adversarial relationship between industry and regulator has its own costs, can itself lead to extremely negative outcomes, and thus has also been justifiably criticized.[12]

But it’s not just about what is the least worst choice.  A close relationship between agency and regulators can provide substantial benefits.  It can improve the information available to an agency, and it can improve regulatory results by increasing compliance and preventing negative unintended consequences.[13]  This is why, in the last few decades, a substantial amount of literature promoted more collaborative government with closer ties and even partnerships between the private sector and regulators.

While capture has benefits on its own, it should not be evaluated in a vacuum.[14]  Rather, we should ask what the realistic alternatives are and which realistic alternative will achieve the best result.  Sometimes capture will achieve better results than other options.[15]  Other times it will at least achieve results that are good enough, or better than, the previous status quo.[16]

The real question is not simply whether “it is good or bad.”  What we should be studying is when would a cozy agency-industry relationship lead to benefits and when would it lead to harm, in light of existing alternatives.  To understand the effects of agency-industry relationship and to make informed policy decisions with that information in mind, we should examine that question theoretically and empirically with as little bias as possible.

This Article makes three contributions to the literature.  First, it makes the point that there is an overlap between the literature on capture and the literature on collaboration, since both address a close relationship between regulator and industry and the results of such a relationship.  It then explicitly analyzes possible distinctions between the two concepts, concluding that the distinction between a situation of capture and a situation of collaboration is a matter of degree or of post-hoc evaluation of the results.  Second, it suggests that a close relationship between industry and regulator has important benefits and describes those benefits.  Finally, it argues that, thus far, the question of which factors increase the potential benefits and which factors increase the risks has been understudied.  This is an empirical question.  A proper study of this question requires theorization and careful examination in real life context.  This Article takes a first stab at answering this question by suggesting such factors.

This Article is a thought piece.  It draws on existing work and current examples to suggest a rethinking of existing scholarship.  It sets out a research agenda for future empirical research.  It does not itself, however, contain new empirical data (though I am in the process of conducting three such empirical studies).

Part I defines the terms capture and collaboration and describes the literature addressing them.  It demonstrates that the terms overlap and there is no clear distinction.  While the literature on capture and collaboration teaches us important insights about the relationship between agencies and industry, a discussion that goes beyond each alone is necessary.  Part II addresses the potential risks and promises of a close relationship between industry and regulator.  Part III suggests factors that can determine whether a close relationship is beneficial or harmful.  The Article then concludes by summarizing the paper’s contribution and explaining avenues for future research.

I.  Capture v. Collaboration

Consider the following two examples of agency-industry interaction.

Since 1975, the FAA has been making use of voluntary reporting programs to allow airlines and airline personnel to report safety problems in return for a guaranteed waiver of sanctions.[17]  Between 1990 and 2009, the FAA entered into agreements for voluntary reporting programs with seventy-three carriers.[18]  In 2001, the FAA promulgated a regulation[19] guaranteeing that it would not use voluntarily submitted safety information for punitive actions, nor would the information be released to third parties under the Freedom of Information Act.[20]  This removed a substantial source of concern for the airlines in relation to the voluntary reporting programs and increased the use of such programs.[21]  Cynics would say this is exactly what industry wanted and is clearly the result of capture.  It allows airlines to get away with violations of safety regulations without sanction because their close connections with the regulators assure them there will be none.

In 2007, a large number of Southwest Airlines planes were found to be in violation of an airworthiness directive,[22] and in some cases, parts were showing fatigue cracks (such cracks themselves are not an immediate problem, but their presence is the first sign of the approach of catastrophic failure).[23]  Apparently, an FAA inspector warned of the cracks as early as 2003, but nothing was done because of the close relationship between Principal Maintenance Inspector Douglas Gawadzinski and senior management at Southwest Airlines.[24]  The connections were so close that FAA personnel allowed Southwest to fly problematic aircraft and even warned Southwest in advance of upcoming inspections.[25]

In April 2011, a tragedy was narrowly averted when a Southwest plane was forced to make an emergency landing because of a catastrophic failure of sections of fuselage skin[26]—the inevitable result of neglected fatigue cracks.[27]  This may be seen as an indication that the FAA and Southwest have returned to their cozy relationship with potentially dangerous results, but this is probably not the case.  Mistakes can be made even without capture, and the National Transportation Safety Board’s (“NTSB”) investigation suggested this was such a case; apparently, none of the parties involved believed the part of the airplane in which the cracks occurred was stressed in a way that would lead to catastrophic failure.[28]  The FAA responded to the near accident by immediately issuing an emergency directive requiring detailed inspections of certain aircraft models when they accumulated 30,000 flight cycles, and thereafter at intervals of 500 flight cycles.[29]  The combination of the NTSB’s findings and FAA’s rapid response suggests a common mistake rather than regulatory failure.

This complex picture suggests that the relationship between the FAA and industry could give critics substantial cause for concern.  But if you take a step back and look at the general picture, what were the effects on air safety?

Allowing the airlines to disclose information without penalty increased the availability of information.[30]  A study by an independent scholar put the total number of reports generated at approximately 900,000 (compared to no avenue to submit such reports before the program was created).[31]  As pointed out by the Government Accountability Office (“GAO”), such data provide information that cannot be found in any other way.[32]  It may have led to fewer punitive measures against airlines, but if the goal was to make air travel safer—not just punish airlines—it seemed to help by increasing the available information.  In a recent piece, Russell Mills, who has been studying these programs for a long time, demonstrated a connection between the voluntary reporting and the FAA’s airworthiness directives, showing that they made a difference (presumably—though it’s hard to measure—for the better) in the regulation of air safety.[33]

An independent review team said the following about the programs: “We are phenomenally impressed with what this agency has achieved, in collaboration with the aviation industry, in driving accident rates down to extraordinarily low levels.”[34]

The rate of fatal accidents among commercial carriers is extremely low, and has dropped dramatically since the 1960s.[35]  While I cannot connect the reduction causally to voluntary reporting—because there were many other factors—the close relationship certainly did not harm, and possibly helped, the situation.

Now consider as a counter the story of the Minerals Management Service (“MMS”).[36]  The MMS received negative attention on two occasions between 2005 and 2010.  In the first case, a lengthy investigation of its Royalty in Kind program exposed a series of ethical problems (described by the Inspector General of the Department of the Interior as a “[c]ulture of [e]thical [f]ailure”).[37]  The MMS collected royalties from companies harvesting minerals from the continental shelf and some federal lands pursuant to leases.[38]  The Royalty in Kind program allowed the agency to receive a percentage of the oil or gas collected by the companies and sell it in lieu of cash royalties.[39]  This became a substantial source of revenue.[40]  The investigation by the Department of the Interior revealed a culture of receiving gifts from industry members—mostly of food, drinks, and lodging (in violation of federal ethics guidelines)—and of alcohol and drug abuse during industry-organized parties (in two cases agency members had to be lodged by the industry because they were too drunk to go home).[41]  There was evidence of relaxation of the rules in favor of the regulated companies, especially in allowing industry members to change bids after winning them.[42]

To complete the picture of a captured agency in thrall to its regulatees, other government officials accused the MMS of lax regulation and carelessness in providing BP with its drilling permit after the BP scandal.[43]  Mary Kendall, Acting Inspector General for the Department of the Interior, criticized MMS regulation as being “heavily reliant on industry to document and accurately report on operations, production and royalties” and pointed out that “[b]ecause MMS relies heavily on the industry that it regulates in so many areas, however, the possibility for, and perception of, undue influence will likely remain.”[44]  The agency was criticized for exempting BP from producing an Environmental Impact Statement—something the agency regularly did for offshore drilling in the Gulf of Mexico.[45]

In spite of this image of a captured agency and the clearly problematic behavior of some of its employees, a recent in-depth empirical examination of the agency’s structure and its history suggests that the accusations of capture obfuscate potentially deeper causes of the problems in the way the MMS functions; many of the problems the MMS ran into are the result of conflicting roles assigned to it, which may have influenced its collaborative approach, and this will not be fixed by the reorganization of the agency.[46]

These two examples serve two purposes.  First, at least some of the behaviors of both agencies can be classified as either collaboration or capture, depending on the observer’s point of view.  Second, they highlight the fact that the kind of close relationship that can be classified as capture can either yield benefits or lead to disastrous results.

A.     Capture in the Literature

Capture focuses on close connections between a regulator and the industry it regulates.[47]  The literature uses a variety of definitions to explain the results or features of capture.  One definition suggests that in a situation of capture, regulated industry members “persuade regulators to alter rules or be lenient in enforcing those rules.”[48]  A somewhat different definition emphasizes the consequences, suggesting that captured regulatory agencies are “persistently serving the interests of regulated industries to the neglect or harm of more general, or ‘public,’ interests. . . . [T]he accusation implies excessive regulated industry influence on regulatory agencies.”[49]

Carpenter and Moss suggest that capture “is the result or process by which regulation (in law or application) is, at least partially, by intent and action of the industry regulated, consistently or repeatedly directed away from the public interest and towards the interests of the regulated industry.”[50]

Braithwaite and Makkai made the capture analysis more nuanced by breaking it into three related but not necessarily simultaneous behaviors—sympathy to industry (implying excessive sympathy), identification with industry’s interest, and (unduly) lax enforcement.[51]

This Article defines capture by emphasizing intentional influence rather than whether the result deviates from the public interest since, at the point of decision, agency decision can often be plausibly seen to be in the public interest—or at least one definition of it; the many aspects of the concept of public interest make a focus on it less than helpful.[52]  It also does not distinguish between influence and control, as Webb Yackee’s article does,[53] since I see the distinction as a matter of degree, a continuum rather than a dichotomy; at some point, influence is so strong as to become control, but in the regulatory context, we are not usually talking about absolute control by industry.

Any of these definitions inherently implies regularly repeated interactions between the regulator and a certain industry.  Capture of the agency is not usually an issue when interaction between the industry and the regulator is sporadic.[54]

This type of relationship has been known for a long time among those inside the regulatory state.  In an open letter about the Interstate Commerce Commission (“ICC”), a famous practitioner acknowledged that:

The Commission, as its functions have now been limited by the courts, is, or can be made, of great use to the railroads.  It satisfies the popular clamor for a government supervision of railroads, at the same time that the supervision is almost entirely nominal.  Further, the older such a commission gets to be, the more inclined it will be found to be to take the business and railroad view of things.  It thus becomes a sort of barrier between the railroad corporations and the people and a sort of protection against hasty and crude legislation hostile to railroad interests.[55]

Research on capture surged in the 1950s, with Marver Bernstein’s influential work on independent commissions.[56]  Bernstein suggested that, while a regulatory regime starts vigorously and with energy, over time its supporting coalition dissolves, the energy dissipates, and the agency starts to kowtow to the preferences of industry, finally seeing its own interests as completely identical to that of the regulated industry.[57]

Bernstein’s work was extremely influential—though it was not without critics from early on.[58]  In the subsequent public policy literature, capture was uniformly seen as a substantial concern.[59]  Even the more “optimistic” studies were optimistic either by suggesting that sometimes agencies avoid capture[60] or by challenging the prevalence of the phenomenon.[61]

In 1971, capture was discovered by the economic literature with Stigler’s seminal article[62] in which he claimed that regulation will, in the end, work for the benefit of the regulated industry, not the public, and that industry will have the most influence on its content.[63]  His approach was given some refinement and algebraic formulation by Peltzman.[64]  The Stigler-Peltzman approach was a challenge to the then-existing conventional wisdom that regulation was justified as a response to market failure, since it was the only way to protect against such failures.[65]  In contrast, Stigler-Peltzman’s findings suggested that regulation was not the answer because it would not work to prevent market failures but would instead reinforce them by giving more power to the regulated industry.[66]  The implied corollary was that market mechanisms are a better solution to market failures.[67]

The Stigler-Peltzman approach achieved prominence and generated substantial follow-up work,[68] though the economic literature on the causes and consequences was mostly theoretical with scant empirical support.[69]  The political science literature on the topic, however, did offer one careful, detailed cross-agency study on the factors leading to capture: Paul Quirk’s 1981 study of the factors increasing the potential for capture.[70]

Much of the literature examined the potential incentives industry can offer regulators to encourage cooperation.[71]  A more specialized strand of literature focused on the revolving door effects, and strongly argued that movement from industry to the regulator and back leads to capture and to regulators wanting to curry favor with industry.[72]

Not exactly fitting into either category, studies in regulation and law found similar results.  Studies of the rulemaking process in the United States found substantive influence by industry on the content of regulation.[73]  In a recent study, Shapiro and Steinzor not only found such influence but explained that it is especially strong where salience (towards the public) is low and technological complexity high.[74]

Most of these studies accept capture as a reality—and a negative one—and discuss what leads to capture and how to prevent it.  Two exceptions stand out.

Recently, Dan Carpenter challenged the existing evidence used in previous studies of capture from Stigler onwards.  In a study that straddled economic and political science literature, Carpenter suggested that deducing capture from regulatory results that seem to favor established firms is problematic since there are other reasons for that kind of advantage, some of which are actually in the public interest.[75]  In other words, you cannot conclude that, just because established firms, large firms, and repeat players do better under the regulatory system, the regulator is captured; there may be other reasons for those results.[76]

In two more studies, regulation scholar Braithwaite and several collaborators went beyond previous literature and suggested that capture may be positive or negative, depending on the circumstances.

In the first study, Ayres and Braithwaite suggested, as part of their discussion of Responsive Regulation,[77] that there are some forms of capture that actually enhance social welfare and are efficient.[78]  For example, one possible effect of capture is that it motivates the agency and the industry to move from a situation in which neither cooperates—a firm evades the law and the agency behaves in a punitive manner—to one where both cooperate—the firm obeys the spirit of the law and works to achieve the regulatory goal, and the agency moves to flexible enforcement, ignoring technical violations.[79]  This situation, explain Ayres and Braithwaite, “is clearly Pareto efficient. . . . [And] will unambiguously increase welfare” since it forces the agency to also consider the firm’s welfare when making decisions—and the firm is part of society.[80]  This analysis is based on theoretical game modeling.

A second study added empirical support to the view that some types of capture are beneficial.  This study, by Makkai and Braithwaite, examined the concept of capture using empirical data from the Australian nursing home industry.[81]  Makkai and Braithwaite suggested that capture should be seen as a more complex concept than was previously suggested, including three factors: sympathy to problems industry confronts, identification with industry, and lack of toughness in enforcement.[82]  They found very limited evidence that ties with industry (coming from industry or returning to industry—the “revolving door” idea) increase capture, and what evidence they found suggested weak effects.[83]

B.     Collaboration

Especially in the past few decades, a whole strand of literature has promoted reforms supporting closer ties between industry and regulators and stronger industry participation in policy making and enforcement.  Much of this literature started as a response to the problems of excessive regulation, command and control models, and conflictual, adversarial relations between agencies and industry; the solution proposed was generally the adoption of more negotiated, consensus-based modes of regulation.[84]

There are two classical examples.  First, in 1982, Bardach and Kagan made a strong, compelling case for the “Good Inspector” acting with forbearance and enforcing regulation with flexibility (and toughness).[85]  Second, in Responsive Regulation, Ayres and Braithwaite argued for a pyramid of enforcement that starts with persuasion and escalates degrees of enforcement on the basis of industry behavior.[86]

Some studies promoted self-regulation by industry as a way to make regulation more efficient and flexible.[87]  Others recommended “public-private partnerships” that would give more of a role to private industry in making policy.[88]  The Endangered Species Act is one example of a program designed to give industry a direct input into the content of regulation.  The Act has been a source of controversy between environmentalists and business interests for a long time.[89]  An amendment to the Act in 1982 allowed the Secretary to approve a Habitat Conservation Plan, allowing a landowner to interfere with an endangered species (“take” it) under certain conditions.[90]  In theory, the option is available to any kind of landowner; however, in practice, timber companies and real-estate developers create most Habitat Conservation Plans.[91]  A Habitat Conservation Plan is, in essence, an official compromise between the private actor (usually a company) and the United States Forestry and Wildlife Service, the agency in charge of implementing the Endangered Species Act.[92]  It is an invitation for businesses to participate in policymaking.[93]  Does this actually protect the environment?  The literature is undecided.  Many studies express concerns and criticisms.[94]  Others express cautious optimism as to the process and its results.[95]

A movement emphasizing the use of Alternative Dispute Resolution tools in enforcement of regulation also emphasizes the need for more cooperation with industry in achieving compliance and less stringent, punitive enforcement modes.[96]  More recently, we see programs that relax enforcement in exchange for voluntary cooperation.  The Environmental Protection Agency has experimented with a number of such programs, with mixed results.[97]  The FAA, as already mentioned, has used voluntary reporting programs with some success.[98]

C.     Lack of Analytical Clarity

While consensus-based regulation and collaboration are not capture, both promote a close relationship between the agency and industry, and the distinction is blurry.  Ayres and Braithwaite openly say that “[t]he very conditions that foster the evolution of cooperation are also the conditions that promote the evolution of capture and indeed corruption.”[99]  Similarly, Shover suggests that “the establishment of cooperative relationships for the solution of problems” is one of the factors leading to capture.[100]

More negatively, an opponent of collaborative government described collaboration as follows: “[T]he very entities subject to regulatory compulsion should engage in the design of rules that will dictate their conduct, self-monitoring for compliance with those rules, and self-enforcement when the entity discovers a violation of those rules.”[101]

Nonetheless, the terms are not supposed to be interchangeable.  What, then, are the differences between collaboration and capture?  Three things stand out.  First, supporters of collaboration want others besides industry involved.  Second, collaboration envisions equal status or the agency as the dominant partner; capture implies the industry is the dominant partner.  Third, collaboration has good consequences; capture has bad consequences.

The first and probably clearest distinction is that most of the pro-collaboration literature does not suggest allowing industry to be the only voice.  In fact, most of this literature wants industry to be a part of a dialogue in which other interest groups are just as well represented and have at least as much influence, if not more.[102]  This potential “countering” role for non-business interest groups is not a new idea, of course.  It was behind some of the moves to more participatory government in the 1960s and1970s,[103] but has since become an integral part of the pro-collaboration literature of recent decades.

For example, Ayres and Braithwaite’s influential Responsive Regulation reserves an important role for Nongovernmental Organizations (“NGOs”) serving as equal participants in the negotiation and enforcement of rules.[104]

For Sabatier too, the solution to capture is participation by third parties, either initiated by the agency or initiated by the interest groups themselves.[105]  Other scholars have also shown positive influence of third-party interest groups as a remedy for capture.[106]

As I will elaborate in Part III, my approach is pragmatic.  I believe in the art of the possible.  In some cases, a third party interest group may be a viable alternative and a good counter to capture (or at least may have the potential to reduce the power of the capturing industry).[107]  In some cases, it will be an excellent option.[108]  However, it will not work in every case.[109]  At least in the American context, there is reason to be skeptical about it working in many cases.[110]

First, it may not always be feasible.  Ayres and Braithwaite believe a suitable NGO—defending the public interest or a relevant private interest—will generally be available.[111]  I am not so sure—and not just in the case of their example of an NGO for the Internal Revenue Service.[112]  In the FAA context, the pilots’ union may be seen as a potential counter to the industry on matters of safety.  However, the pilots’ union’s interests on the issue are mixed since it also has an interest in increasing jobs and the profitability of the industry, so it does not quite represent the public interest.  Further, consider the MMS.  In that context, what NGO promotes “honesty in dealing with the oil industry”?  Environmental organizations may be watchdogs for compliance with environmental requirements, but what about the rest?  For many regulatory issues, the problem of diffused interests mentioned in the economic literature is all too real.[113]  Furthermore, being at the negotiating table requires resources—even if all information the regulator has is made available,[114] the NGO will still have to invest in processing the information and involvement—especially in more complex regulatory areas, where there are many decisions.  There may not always be a group with the required dedication or the ability to constantly engage.  We can learn something about this from the United States’ notice and comment rulemaking process which provides an opportunity to any interested party to participate, at least to some degree.[115]  In spite of the apparently open access, many studies suggest that industry participates more than others in this process.[116]  The same reasons that industry captures agencies operate here too.  Industry has the most interest in the content of regulation, and it is a long-term, substantive interest, which keeps industry involved after other participants have left the field.[117]  Industry has substantial resources to invest in affecting the content of regulation, and strong incentives to do so.[118]  In some situations, there will be an NGO with the same dedication and the ability to persevere; in others there will not.

Furthermore, even where such an NGO is available, it will not necessarily solve the problem.  One question is who will guard the guardian: how do we prevent the capture of the NGO by the regulated industry?[119]  Ayres and Braithwaite suggest as the solution a contested representation.[120]  This, however, requires not one but multiple suitable NGOs able and willing to undertake the role.

Similarly, if an NGO exists that is willing to counter the influence of the capturing industry it may improve the regulatory result—or not.  It all depends on which interests it represents.  Not all clashes between private interests protect the public interest.  Just as important, “[e]verything has a price.”[121]  Allowing a third party to counteract the influence of industry carries the cost of added delays, and depending on the level of trust between actors, can cause a return to a more contentious process.  It thus has the potential to thwart the original intent of the collaboration altogether.  Sometimes the price will be worth paying.  Other times, if the results of capture are good enough and the level of conflict high enough, it will not be.

A second distinction between collaboration and capture is that the term collaboration does not imply, as capture does, that the dominant partner is industry.  It implies a partnership of equals.  Calling a regulatory relationship capture suggests that industry usually gets its way, often because the agency really buys into the industry’s views.[122]  This is an important difference, but it seems more a matter of degree than of kind; how much real influence does industry have in practice?  If an agency accepted industry’s view, was it captured, or was it convinced by industry’s arguments?  Determining the degree of industry’s influence may be difficult in practice.  And even where that is possible, the argument still stands that mechanisms of collaboration facilitate and create opportunities for capture.

The final difference is a difference in results.  Calling a relationship between industry and regulator collaboration implies a positive outcome; capture implies a negative one.  But evaluation criteria based solely on results are not very useful as a conceptual tool to identify capture or other phenomena.  Results, obviously, can only be examined in hindsight.  In terms of normative judgment, they do not help assess the behavior itself.  Results do not provide a very useful guide, especially since it’s not so obvious we can identify capture when we see it.  For starters, identifying whether regulation works for the benefit of a certain industry is not easy.  In an early study, Etzioni challenged Stigler and Friedland’s early finding that utilities had captured regulators[123] by using their original data to arrive at the alternative conclusion—that regulation actually benefited consumers and therefore the regulators were not captured.[124]

Even if regulation does work for the regulated industry, it is not at all clear that capture is at work.  In an innovative piece, Carpenter demonstrated that there are other reasons besides capture that regulation may benefit large, established firms.[125]  These firms have real advantages.  For example, they are more known to the regulator, who can therefore make quicker decisions in relation to them, may enter niches earlier that are politically valuable, and can better withstand delays in decision making.[126]

Besides the difficulty of identifying capture, given the negative implications of the term, if used as a basis for policy prescriptions, it is problematic to say to the agency after the fact, “at first we thought you were collaborating and all was well, so we encouraged you to keep at it, but now we see that, since things turned out badly, you were in fact captured, and therefore there will be consequences.”[127]

Both capture and collaboration address situations of a close relationship between agency and industry.  The analytical distinction between them is problematic, and even when it can be used, collaboration mechanisms can be seen at least as “capture facilitators.”  The question is, therefore, what are the possible consequences of such a close relationship?

II.  The Risks and Benefits of a Close Relationship

A.     The Risks

The risks of a close relationship between agency and industry are that it could lead to weak (or absent) protection of the public, result in benefits to the industry at the expense of the public (“rent-seeking”),[128] and lead to straight out corruption.

One common justification for regulation is protecting the environment, public health, or other important values against economic interests of private firms.[129]  When it comes to protecting other important values (such as health, safety, or the environment), industry members naturally want to minimize their costs in order to increase their profits; in fact, it can be argued that they have a duty (to their shareholders) to do so.[130]  One result of this, goes the argument, is that industry’s influence will lead to ineffective, weak, and watered-down regulations that, in fact, do not provide adequate protections in these areas.  For example, one study suggests that, because of industry influence, the Food Safety and Inspection Service in the United States Department of Agriculture (“USDA”) had done nothing in relation to E. Coli-contaminated meat for over a decade (1982–1995).[131]  This neglect continued in spite of repeated outbreaks of illnesses related to E. Coli and after a report by the National Academy of Sciences, which pointed out the shortcomings of the USDA’s method of inspection and suggested a different one.[132]  The situation only changed after a very widespread outbreak of E. Coli, which was traced to contaminated meat and had dramatic and obvious effects (including the deaths of two children).[133]  Even then, the final rule was a watered-down version of the original, with most of the effective protections removed—a result of intensive efforts by industry.  Industry wanted a weak rule, and it succeeded in getting just that.[134]

Similarly, Etzioni suggests that a rule by the Department of Transportation was “profoundly shaped” by the railroad industry.[135]  The Department of Transportation allowed railroads to choose the routes acceptable for dangerous cargoes based on their own weighing of the factors involved (including the costs).[136]  Etzioni discusses a claim by Melberth from OMBWatch that the lobbyists from the industry actually provided the text of the rule.[137]  This rule, according to critics, failed to require rerouting of dangerous cargoes around major cities—including those designated by the Department of Homeland Security as targets for future terrorist attacks—with the result that the railroads continued to route such cargoes through population centers, potentially endangering lives.[138]

Etzioni offers another example.  In 2000, when the FDA prepared to publish information about the mercury content of various foods, the tuna industry—realizing canned tuna was going to be classed as dangerous—lobbied the FDA.[139]  The FDA recalibrated its categorization, and an FDA official, Clark Carrington, admitted that the staffers designed the three categories of mercury danger so that canned tuna fell into the “low” category to “keep the market share at a reasonable level.”[140]

Furthermore, capture can also lead to complete removal of regulations—deregulation—a phenomenon described by Carpenter as “corrosive capture.”[141]

The risks of a close relationship between agency and industry are also found in relation to enforcement.  Due to the close relationship between the regulators and the regulated companies, the regulators will be unlikely to do their job and rigorously enforce the regulations.[142]  This will lead to decreased protections for the public.  The regulators will be unwilling to penalize their good friends in the industry.  On the contrary, they may seek to please and promote those to whom they are personally connected.  After all, Great Walls can always be brought down by great lunches; in other words, the separation into regulated and regulator may not survive close personal contacts.  This is especially true where there is an exchange of personnel and close interpersonal connections between agency officials and industry employees.  Indeed, much of the economic literature about capture focuses on the negative effects of the “revolving door”—people moving between industry and the regulators.[143]  This tendency was described by the Department of the Interior’s Acting Inspector General, Mary L. Kendall, as part of the reason for the ethical problems discovered in the MMS, where there was “a culture where the acceptance of gifts from oil and gas companies were [sic] widespread throughout that office.”[144]

Of greatest concern to me is the environment in which these inspectors operate—particularly the ease with which they move between industry and government. . . . [W]e discovered that the individuals involved in the fraternizing and gift exchange—both government and industry—have often known one another since childhood.  Their relationships were formed well before they took their jobs with industry or government.[145]

Even without corruption, close connections can foster excessive trust that will lead to accepting the words of the industry at face value and, therefore, not finding out about violations.  For example, in relation to the FAA, a report by the Nader group found that FAA inspectors often missed industry violations because they relied on paperwork provided by the industry to discover problems, and the industry would lie in writing—a practice termed “pencil whipping.”[146]

In another example, the Bureau of Land Management (“BLM”), the agency handling on-shore oil drilling, was found by the GAO to regularly approve drilling permits without an Environmental Impact Statement, relying on exclusions provided in section 390 of the Energy Policy Act.[147]  The GAO found that the MMS (acting under the aegis of the BLM) used such exclusions in more than a quarter of drilling permits between 2006 and 2008,[148] frequently “out of compliance with both the law and BLM’s implementing guidance.”[149]

Preparing an Environmental Impact Statement, as required by the National Environmental Policy Act (“NEPA”),[150] is costly and difficult, and removing the requirement is a break for industry, but it raises concerns about adequately identifying environmental consequences and thus environmental protection.  In a specific example, the MMS (at the time under the BLM) apparently granted an exclusion from the NEPA to BP’s drilling in the Gulf of Mexico—drilling that ended with a catastrophic, far-reaching oil spill.[151]  The natural accusation is that cozy relationships between the agency and the company led to lax enforcement of the legal requirements, sacrificing the environment to the company’s interest.  According to a member of an environmental group quoted in the article, “[t]he agency’s oversight role has devolved to little more than rubber-stamping British Petroleum’s self-serving drilling plans.”[152]

Another common justification for tight regulation is that it can correct market failures (e.g., by limiting monopolies and cartels).[153]  The situation here is somewhat different.  In a monopoly situation, industry will want to place barriers on new entrants to the market and make access difficult.[154]  For example, one of the common struggles in relation to telecommunications liberalization is to allow potential new entrants access to the existing network to prevent them from having to invest the tremendous costs of creating a network anew.[155]  Research found that opening the market to competition required regulation to prevent the incumbent from setting access prices too high, and thus abusing their market power.[156]  Close connections between the incumbent and the regulator can lead to setting the access prices high.[157]

When it comes to rate setting, established industry will want a lax regime that provides it with maximum freedom.  For example, in his study of railway prices, Huntington demonstrated that, since the late 1920s, the ICC, captured by railroads, consented to any rate increase the railroad wanted.[158]

B.     The Benefits

While the risks of a close relationship are real enough, industry involvement in writing regulations is a recurring phenomenon,[159] as is flexible (or lax) enforcement.  Many studies have found that regulators rarely act punitively and generally prefer to negotiate and work with industry rather than prosecute or punish it.[160]  And that is not just because of the problems of the administrative state; there are good reasons to want industry involvement in the creation and enforcement of regulation, in spite of the obvious risks.

The first is that the best information about what is going on in industry is found in the hands of industry.  The second is that working with industry can lead to better results; enforcing compliance, without cooperation, is costly and, at best, only partly efficient—because, among other things, an industry unwilling to cooperate can find many ways to obstruct or avoid enforcement.[161]  The third is that it can also lead to better results because there is a societal advantage in preventing unanticipated negative consequences for industry,[162] and industry is often in the best position to anticipate and warn against such consequences.

1.     Industry and the Information Advantage

Good regulation requires good information.  Not only is this self-evident, but the legal framework is designed to increase the information available to agencies.[163]  At least one goal of the notice and comment process the Administrative Procedure Act (“APA”) mandates for informal rulemaking is to provide information.[164]  Many of the other requirements added to the process require the agency to undertake research that will make its decision more informed.[165]

A serious problem with this process is the fact that often the best information about what is going on in a given industry is in the hands of members of that industry.  Agencies are regularly understaffed and overworked[166]—a reality that is getting worse in these days of budget deficits and in this “age of austerity”[167] when agency resources are constrained and reduced.[168]  They cannot, even with the best will, collect all the needed information about industry to either devise the best regulations or catch those who violate them.  For example, the FAA is, on its face, a large agency with over 50,000 employees.[169]  But many of these employees are air traffic controllers.[170]  There are only about 4000 inspectors overseeing all the planes in the United States.[171]  As a second example, the Food Safety and Inspection Service of the USDA has to oversee the safety of meat and poultry in the United States.  The growth in meat consumption and sale has left its inspectors overburdened; one author estimates that “[a]t ninety-one birds per minute, inspectors have to examine over 12,000 poultry carcasses each day.  It is estimated that inspectors have an average of just two seconds to inspect a poultry carcass and twenty to thirty seconds to examine a 2,000 to 3,000 pound beef carcass.”[172]  A similar claim has been made concerning FDA resources.[173]

It is easy to claim that the solution should be for Congress to provide more funding,[174] but the chances of that happening are not very high in a political climate that calls for reducing government size and limiting spending.[175]

Along the same lines, producing information costs money.  And in an era of budget cuts, the agency often has to choose between producing the information and doing other things.  On the other hand, industry often needs the information in question for other purposes (e.g., information on safety problems in airplanes) and may be collecting it anyway.  Getting the information from industry can save money.  But to whom will industry be more likely to provide the information: the cop who monitors it and is prepared to punish it or the friendly regulator who goes out to lunch with it?

Most importantly, even if an agency had all the personnel and funding it could wish for, its information would still be secondhand (or we might say thirdhand since the information inspectors collect still needs to go up the agency hierarchy and be processed before it reaches central decision makers).[176]  It is the industry people who work on the ground and know what is really happening.[177]  They have the best opportunities to spot problems, and they can be the first line of defense.[178]  The problem, of course, is that industry members will have no incentive to provide information that might later hurt them.  They will have even less incentive to provide such information in an adversarial, punitive, hostile environment.[179]  If the relationship with the agency is good, and especially if industry members believe regulators care about industry’s interests, industry will have more incentive to provide information and will be more willing to trust the agency with it and to try and convince the agency in noncontentious ways of its point of view.[180]

Of course, another solution to the “not having good enough information” problem is requiring industry to provide such information and heavily punishing any parties that withhold it.  However, that may backfire.  Industry can respond by providing too much information.[181]  Information, even when provided, needs to be processed and considered.  In fact, one of the most common problems in the modern world is the problem of “info glut”—having too much information.[182]  Sorting through information also requires resources; extracting nuggets of meaning from a mass of verbal gravel can be very labor intensive when information is not well presented.[183]  Close relationships can reduce the motivation to practice this sort of information dumping and can incentivize industry to provide the information in a more useable form.[184]

Information overload is not the only potential risk of coercive or punitive information gathering.  As with any other issue, achieving compliance through coercion is neither simple nor straightforward.  Getting the information voluntarily through cooperation is often more efficient and easier.[185]  There is at least one study that suggests it may incentivize industry to conduct more research.[186]

If industry has a direct involvement in writing regulations, the regulation may be self-serving and weaker than it might otherwise be, but it will probably be well informed.  That will help prevent ineffective or erroneous regulation that may have substantial unintended consequences.

At the same time, relying on the regulated industry for information raises at least two real problems.  First, industry will probably provide information that supports its interests, place emphasis on things that support its views, or tend to downplay the things it prefers not to have regulated.  Industry may even do that without intending to; a known cognitive bias is the confirmation bias, which suggests that people (or companies) tend to emphasize and be more receptive to things that support their initial point of view.[187]  Almost automatically, the tendency will be to downplay or ignore adverse information—to rationalize it away.[188]  This is not true just of industry.  Sabatier pointed to the tendency of advocacy coalitions to “resist” information that suggests their core beliefs are wrong or their preferred outcomes unattainable, and to embrace information that supports their preferred point of view.[189]  Thus, simply by the nature of things, industry will tend to believe in, and provide, information that represents its interest in the best available light.  Close connections do not reduce the risk of self-serving information—but neither does the absence of such connections.  And the absence of information from industry can sometimes be much more costly than the receipt of self-serving information, as in the example of the FAA’s voluntary disclosure programs.

Another danger is that industry will lie.[190]  One could argue that a close relationship might increase the risk of lying because the industry will expect the agency to believe it and therefore expect the chances of getting caught to go down.  But the risk of lying exists regardless of the relationship with the agency.  One could equally argue to the contrary—that the industry will be more likely to lie (and will be more inclined to feel that lies are justified) if the government is “the enemy”[191] than if the relationship is good, and especially if industry expects the regulator to have a realistic regard for its legitimate interests.

A more subtle, but just as real, danger involved in capture is the effect of trust on the testing of information.  The problem here is that if the agency and industry have close connections, the agency may see information provided by its good friends in industry as reliable and not make adequate efforts to confirm or verify it.  Mistakes can happen even with a completely transparent, cooperative, and honest industry, and thus verifying information is useful and important.  In fact, for anything but perfection, it is crucial.  For example, the “pencil whipping” described by Nader and Smith can be traced in part to the FAA’s trust that the airlines provided reliable information, when they clearly did not.[192]  Similarly, another Nader report—this time on the FDA—suggests that, in relation to food additives, the FDA bought into industry assumptions and accepted some self-serving and misleading research results without doing its own testing.[193]  On the other hand, this kind of thing is not limited to close regulatory relationships.  Even an agency that is not “captured” will have trouble carefully scrutinizing data provided by industry.  Staffing problems and lack of inside information means agencies do not—cannot—test most of the information they get from industry.  Instead, they often rely on self-reporting.  A close relationship can improve the quality of self-reporting since relationships work both ways; the industry, too, will not want to disappoint its regulator allies or jeopardize the connection.

At the end, it comes down to a question of which problem one would rather face.  Is it better to have more information, at the risk of that information being self-serving or even unreliable, or is it better to lack information and make mistakes because of that?  If regulation is the art of the possible, lack of information makes very little possible.  The information provided by industry may well be partial and self-serving, but it is more than the agency will have in a more conflictual scenario.

2.     Improving Regulatory Results: Compliance

A close relationship—up to the level of capture—can also improve regulatory results by improving compliance.  Thoughtful students of regulation demonstrated that beyond a certain point, strict enforcement can backfire and lead to less, not more, compliance.[194]  The reason is that it can create resentment, which will lead to resistance and passive compliance.[195]  That is not to say that having strict sanctions is not useful, but it is useful most of all as a tool of last resort and as a background for strategies of negotiation and cooperation.[196]

Many studies have demonstrated the limits and problems of coercive enforcement.  A scholar of regulation recently said:

Research on second-generation regulatory agencies made clear to many that there are inherent shortcomings and limitations of strict rule-based enforcement.  Investigators from Australia to the U.S. found that despite the content of regulations, oversight of business firms was exceedingly imperfect. . . .  It was obvious to a host of investigators that a rigorous deterrence-based approach to oversight of privileged offenders that the level of political and fiscal resources it would require may [sic] it unlikely if not impossible.[197]

Studies suggest that many agencies do not use punitive sanctions even when they are available.[198]  And if they do use punitive sanctions regularly, the costs—not just the direct costs, but the negative effects—can be very high.[199]

Voluntary compliance is, by many standards, better than punitive enforcement.  It is cheaper—the agency does not have to invest as much in monitoring and in prosecuting wrongdoers.[200]  Punitive actions have costs in terms of personnel and time.  In the United States, many actions against wrongdoers also involve the courts,[201] which is another consideration.  Litigation is expensive.[202]  Even more importantly, quite often to achieve the regulatory results, you need the regulated firms to be willing to make an effort and occasionally suggest creative solutions.[203]  Even if it is not strictly necessary, industry creative involvement can lead to more efficient solutions.[204]  An adversarial relationship will discourage such behavior while a positive one will promote it.  Not to mention that punitive enforcement requires appropriate rules to be specified in advance—and in complex modern realities, regulations are almost always going to be over or under inclusive.  Cooperation by industry with the goal of the regulation can achieve better results.[205]

The side effects are less negative with voluntary compliance than with strict punitive enforcement.  Bardach and Kagan demonstrated the pitfalls of strict punitive enforcement: creating resentment on the part of the regulated, which see enforcement as arbitrary;[206] leading the regulated companies, even the “good apples” among them, to just do what the regulator demands and not invest in additional responsible behavior;[207] undermining cooperative problem solving;[208] and possibly leading the regulated industry to give up on the rule of law.[209]  Many subsequent studies suggested that punitive enforcement can harm cooperation.[210]

Close relations between industry and regulator—and influence of industry on the regulator’s view of sanctions—certainly reduce aggressive enforcement or even enforcement in general.[211]  But they can also increase voluntary compliance—at a price, and with an attendant risk.  If industry is involved in writing the regulations, it presumably agrees to the content and can be expected to comply with that content.  If enforcement is flexible and negotiated with the regulated industry—if it is done by agreement instead of by fiat—it is more likely the agreed-upon modifications will be put in place.

The price is sacrificing some of the results that could be achieved by top-down regulation.  If industry is going to agree to a regulatory scheme limiting it or imposing costs on it, it will probably agree to less than the supporters of the regulation want, to what can be seen as watered-down regulation.  In some circumstances, the regulation may be watered down to such a degree that the result will be the sacrifice of the other values.  But it does not have to be.  Industry has other reasons for not going too far in weakening regulation.  Among other things is the concern of reaction.  If regulation is too weak, there may be a public backlash, at least if there is a bad result.  The harsh reaction to the perceived cozy relationships between FAA and Southwest described in Part I is one example; the pressure led the FAA to substantially increase its enforcement, possibly too much.  In the following days, it grounded large numbers of American Airlines planes too, disrupting travel.[212]  Concerns about regulation have been a reason for industry to self-regulate.[213]  The same logic can lead industry to support a higher level of regulation than it would absent any outside pressures: better to have a hand in the process and cooperate (and be seen to cooperate) than to have the regulations imposed.

Furthermore, in many circumstances, industry will be at least partly on board with the goal of regulation.  In terms of the FAA, quite a bit of support for voluntary reporting programs came from the pilots working for the allegedly capturing airlines.[214]  Obviously, safety is also a major interest of the pilots and staff on the planes, who have constant exposure to whatever hazards exist.  A similar point is true for nuclear plants: a nuclear explosion endangers not only the public but also everyone in the plant, and makes it harder to get a permit to build another plant or fix things, when needed.[215]  And after all, industry is not one skin; top management may have an interest in avoiding scandals—be they airplane crashes, fatalities because of negligently manufactured drugs (such as when Chinese manufacturers deliberately used a cheap substitute instead of dried pig intestines to make the drug Heparin, leading to eighty-one deaths),[216] or nuclear explosions.

Other studies suggest that at least some corporations can have real commitment to social values, such as the environment.  That is one justification for voluntary compliance programs, some of which focus on “high performers.”[217]  From the point of view of the more socially conscious companies (the “knights,” drawing on Le Grand’s terminology),[218] the absence of regulation may disadvantage them since it allows less conscientious companies to cut corners and thus produce products more cheaply.  Those companies may have an interest in pushing for stricter regulation.[219]

Not only might the sacrifice be less than anticipated; there is, again, a question of “the art of the possible.”  Sometimes it is better to compromise on the content of the policy and end up with a policy that is easier to implement and more workable in practice than have a stronger policy that does not work.  And as long as the goal of regulation is to achieve results and not just punish industry for being industry or for being big industry, a compromise that achieves something may not only be the best possible solution under the circumstances, but it is itself a positive thing.  After all, strict enforcement or more adversarial mechanisms do not usually achieve one hundred percent compliance either.[220]

3.     Capture and Regulatory Results: Unintended Consequences

Regulation is not often planned with the intent to harm an industry.[221]  For example, in spite of views to the contrary, the goal of most environmental regulation is not to destroy any branch of industry or put workers out of jobs.  It is to protect the environment.  Unfortunately, sometimes the precautions needed to protect the environment, public health, competition, or any of the other things regulation tries to achieve are costly.  They can be costly in terms of direct monetary costs to the industry,[222] or in terms of reducing industry’s competitive edge compared to industry in other countries.[223]  They can be costly in other ways; for example, they can work to the advantage of large companies and against small businesses, thus pushing a sector towards oligopoly.[224]  They can have costs in unanticipated directions, such as delay on large construction projects.[225]

Strong involvement of industry in creating the regulation and allowing it substantial influence on the way regulation is enforced takes seriously concerns about negative effects on industry.  When designing regulation, industry can warn agencies in advance of potential costs and work with them to mitigate such costs.  Industry can negotiate enforcement that will not lead to unintended consequences.

At the same time, two risks are attendant.  The first is that, while capture allows us to take seriously the risks to industry from a certain kind of regulation, it may not give the same weight to unintended consequences for other groups, such as low-income people or small business.  The second is that capture may take such account of the risks to industry that the attendant regulation will not protect other important values—they will have no “bite.”

Again, the challenge is one of achieving maximum results with minimum sacrifice.

III.  Discussion: Factors Affecting the Results of Capture

Part II suggests that a close relationship between industry and regulator—up to the level of capture—carries risks and can result in very bad consequences, but also has important potential benefits.  In fact, some degree of close relationship may be necessary for a functional relationship between regulator and regulated industry.

My approach is pragmatic.  Regulation aims to achieve specific goals.[226]  Regulation is not in place to decorate shelves with rulebooks (or not only; leather-bound rulebooks certainly add to a room’s atmosphere).  Therefore, most discussions surrounding regulations rightly focus on what the results should be and how to achieve them.  If capture can, in certain circumstances, promote those goals—or promote them better than other tools—it should be used for that goal.

In that vein, when looking at the relationship between industry and regulator, we should focus on how to maximize its positive results and minimize its dangers.  To some extent, that is an empirical question, but at least some factors can be suggested.  As a starting point, the benefits highlighted in this Article are information, improved compliance, and avoiding unintended consequences.  The most important factors will relate to those benefits.

A.     Information

The importance of industry-only information is significant here.  A close relationship will be more beneficial where information is really difficult or expensive to come by without industry cooperation.  The problem is that the risks of a close relationship—because of the difficulty of verifying the information—are also extremely real in this situation.  If industry knows its information cannot be verified, and especially if there are other pressures, it may be tempted to massage such information.  This may be a situation where a close relationship between industry and regulator is essential, but close external supervision of the regulator—or at least occasional close scrutiny—is a necessary corollary.  At least some literature suggests that monitoring itself can prevent abuse.[227]

A stronger factor is where within industry information is available.  Industry is not one skin.  While the question needs empirical investigation, I suggest that the benefits will be higher where the information an agency needs is in the hands of the lowest and highest echelons of industry, in contrast to middle managers.  Lowest echelons may have less incentive to hide information from agency and, if the relationship is good, may be closer to the regulator than to management.  They are also the ones that may suffer from some problems—such as airline accidents.  Highest echelons may buy into the regulatory goals, and at any rate, have much to lose from scandals.[228]  But middle managers, who are often under substantial pressure to get results and get things done, may find themselves cutting corners more often and may wish to hide those cut corners from both management and regulators.

Second, while the risks do exist, the benefits are higher when understanding the way industry works “on the ground” is critical to effective regulation.  As pointed out by Makkai and Braithwaite:

[I]nspectors who come from the industry bring with them not only some special insight into the difficulties the industry faces, but they also bring special insight into the tricks of the trade used to get out of those difficulties.  Industry experience can be helpful in finding the skeletons in the corporate closet.  Admittedly, inspectors take the tricks of the regulatory trade across to the industry as well.  But it is clearly the government that gets the better of this particular exchange.  This is because most of the regulator’s job involves dealing with industry, while only a little of the business person’s job will concern dealing with regulators, unless she becomes a regulatory affairs specialist in a large firm.[229]

B.     Compliance

The incentives for industry to comply with regulation are also an important factor.  Noncompliance can be costly for industry.  One important factor is the existence of potential victims of noncompliance.  Potential victims of noncompliance who can complain or sue increase the risk of detection and of negative reactions, and therefore increase industry incentive to comply with regulation.  Deaths from airplane accidents, children harmed by specific products, and people who lost their homes following the mortgage crisis are more likely to generate sympathy and lead to outrage than harm to the general taxpayers’ base.

The benefits to the industry from the regulatory regime are also important; does regulation help coordinate between parts of industry?  Do regulatory requirements help management achieve values it already wants to achieve?

Finally, there is a question of what is on the other side of the scales.  If the costs of compliance are really high, the best relationship in the world may not push industry to comply.  So if the costs of compliance—direct or indirect—are substantial, the risks of capture are more likely to materialize.

Second, the incentives for an agency to stay true to its mandate are also important, and the salience of the regulated industry matters here.  A less-noticed agency may feel more comfortable allowing industry to deviate from the public interest than one that is under scrutiny.  The FAA regularly receives critical media attention after air crashes, leading some commentators to see it as a “tombstone agency”—an agency that only acts when someone dies (and in proportion to the number of deaths).[230]  The negative attention puts pressure on both the FAA and the airline industry, criticized together, to act.  In that sense, a close relationship may be a benefit.  When the pressure comes from outside the agency and is directed at both agency and industry, the close connections may make acting more effective: “We have a common problem.  Let’s solve it.”

The MMS, on the other hand, was hidden from the view of anyone outside Washington or oil companies until the scandals related to it occurred (the accusations of corruption due to gift taking by its employees and the accusations of lack of enforcement of environmental regulation in connection to the BP oil spill); even now, most people will have difficulty recognizing its name.  With no external exposure, there was no reason for the agency officials identifying with industry to beware or act differently.  In relation to the Royalty in Kind program, there came a point where the rest of the administrative state caught on and stepped in to correct the problem.[231]

Third, we have to ask what the real-world alternatives to capture actually are.  If an agency has enough funding and aid from external constituencies or from other sources in enforcing rules, or is facing a very divided industry, it may get a great deal done without capture.  In that case, the risks of capture may stand out more (though even there, the benefits are important).  But if the agency is substantially understaffed for its assigned role, or underfunded, capture may be the only way to get anything done.  In that case, achieving what an industry is willing to give through cooperation and due to its good relationship with the regulator may be enough, or at least better than nothing.  Capture will provide more benefits if there is a credible possibility of sanctions in the background.

In other circumstances, alternatives to capture may exist—command and control regulation, or a very comprehensive process, may be possible—but capture will still derive more benefits, or derive them more cheaply (and so may be better).

In the words of Komesar:

The correct question is whether, in any given setting, the market is better or worse than its available alternatives or the political process is better or worse than its available alternatives.  Whether, in the abstract, either the market or the political process is good or bad at something is irrelevant.  Issues at which an institution, in the abstract, may be good may not need that institution because one of the alternative institutions may be even better.  In turn, tasks that strain the abilities of an institution may wisely be assigned to it anyway if the alternatives are even worse.[232]

If an institution works well, but others would work better, others should be used; if an institution has problems, but the alternatives are even worse, that institution is the best possible for the specific context.[233]  In this context, capture, in spite of its manifest drawbacks, can sometimes be the best alternative we have available.  Thus, before finding fault, any critic must address the issue of what better course might have been taken.


Concerns about capture are still very real—though, these days, the belief in the value of collaboration provides something of a counterargument.  This Article suggests a way to reconcile the extensive literature denouncing the dangers of capture with the literature emphasizing the benefits of collaboration by suggesting that indeed, there is an overlap—at least a potential overlap—between them since both highlight a close relationship between industry and agency, but that overlap is more than just a cause of concern. Such a relationship has its benefits.

The devil is, as always, in the details—what is the relationship between the regulator and the industry, who else is on the playing field, what are we trying to achieve, and what else is available?  These are empirical questions.  A blanket condemnation of close relationships between industry and regulators by naming them capture is problematic and can lead to sacrificing potential advantages.

We need to start empirically studying and evaluating the factors that make a close relationship between industry and regulators work or vice versa.  We need to tackle the formidable task of assessing its real effects on public policy.  It is formidable because it requires defining what the “public interest” is in the relevant area and assessing the effects on it, neither easy tasks.  But it is important.

         * Associate Professor, UC Hastings College of the Law.  I would like to thank Mark Aaronson, Hadar Aviram, Ash Bhagwat, Abe Cable, Daniel Carpenter, Marsha Cohen, David Coolidge, John Crawford, Annie Daher, Ben Depoorter, Christopher Elmendorf, Heather Field, Beth Hillman, Alicia Jovais, Robert Kagan, Chimene Keitner, Heather Kelly, Evan Lee, Ethan Leib, John Leshy, David Levine, Richard Marcus, Jerry Mashaw, Osagie Obasogie, Roger Park, Radhika Rao, Reuel Schiller, David Schorr, Darien Shanske, Gail Silverstein, Brendon Swedlow, David Takacs, and Joan Williams for invaluable comments and suggestions in earlier stages of the project and on earlier drafts, and Erica Anderson and Fatemeh Shahangian for excellent research assistance.  All errors are, of course, my own.

        [1].   Toni Makkai & John Braithwaite, In and Out of the Revolving Door: Making Sense of Regulatory Capture, 12 J. Pub. Pol’y 61, 72–73 (1992).

        [2].   Scholars have also discussed concerns about capture by Nongovernmental Organizations (“NGOs”) and other interest groups.  See, e.g., Dieter Helm,Regulatory Reform, Capture, and the Regulatory Burden, 22 Oxford Rev. Econ. Pol’y 169, 172–75 (2006); Richard L. Revesz & Allison L. Westfahl Kong,Regulatory Change and Optimal Transition Relief, 105 Nw. U. L. Rev. 1581, 1604–07 (2011).  However, most of the scholarship focuses on the connection between industry and the regulator, and the case studies explore that situation.  I will therefore use language focusing on industry, while acknowledging that there can be capture by other actors.

        [3].   Capture has been discussed at least since the 1950s.  See Marver H.  Bernstein, Regulating Business by Independent Commission 268–71, 277–79 (1955).  Kenneth Culp Davis describes Bernstein’s work as “[t]ypical of the prevailing attitude among the present generation of political scientists.”  1 Kenneth Culp Davis, Administrative Law Treatise § 1.03, at 16 n.2 (1958).  It is still a topic of discussion today.  See, e.g., Steven P. Croley, Theories of Regulation: Incorporating the Administrative Process, 98 Colum. L. Rev. 1, 1–7 (1998); Amitai Etzioni, The Capture Theory of Regulations—Revisited, 46 Soc’y 319, 319–20 (2009); Michael E. Levine & Jennifer L. Forrence, Regulatory Capture, Public Interest, and the Public Agenda: Toward a Synthesis, 6 J.L. Econ. & Org. 167, 178–79 (1990); Makkai & Braithwaite, supra note 1, at 62; Sidney A. Shapiro & Rena Steinzor, Capture, Accountability, and Regulatory Metrics, 86 Tex. L. Rev. 1741, 1742, 1745–46, 1756, 1759 (2008); Wendy E. Wagner, Administrative Law, Filter Failure, and Information Capture, 59 Duke L.J. 1321, 1321, 1328–34 (2010).

        [4].   Mark Bovens, Public Accountability, in The Oxford Handbook of Public Management 182, 182 (Ewan Ferlie et al. eds., 2005) (“As a concept, however, ‘public accountability’ is rather elusive.  It is a hurrah-word, like ‘learning,’ ‘responsibility,’ or ‘solidarity’—nobody can be against it.”).

        [5].   See infra notes 60–61 and accompanying text for a discussion of the more “optimistic” strands of literature about capture.

        [6].   See infra Part I.A for a more thorough definition of capture.

        [7].   Martha Chamallas, The Disappearing Consumer, Cognitive Bias and Tort Law, 6 Roger Williams U. L. Rev. 9, 18 (2000).

        [8].   James S. Turner, The Chemical Feast: The Ralph Nader Study Group Report on Food Protection and the Food and Drug Administration 18 (1970).

        [9].   Id. at 37.

      [10].   Id.

      [11].   Ralph Nader & Wesley J. Smith, Collision Course: The Truth About Airline Safety 83–89 (1994).

      [12].   Eugene Bardach & Robert A. Kagan, Going by the Book: The Problem of Regulatory Unreasonableness 102–19 (1982); Robert A. Kagan, Adversarial Legalism: The American Way of Law 195–98 (2001) [hereinafter Kagan, Adversarial Legalism]; George J. Busenberg, Collaborative and Adversarial Analysis in Environmental Policy, 32 Pol’y Sci. 1, 1–2 (1999).

      [13].   Russell W. Mills, The Promise of Collaborative Voluntary Partnerships: Lessons from the Federal Aviation Administration 14–16 (2010).

      [14].   See generally Neil K. Komesar, Imperfect Alternatives: Choosing Institutions in Law, Economics, and Public Policy 3–7 (1994) (arguing for a comparative institutional analysis); Jeb Barnes, In Defense of Asbestos Tort Litigation: Rethinking Legal Process Analysis in a World of Uncertainty, Second Bests, and Shared Policy-Making Responsibility, 34 Law & Soc. Inquiry 5, 6–7, 11 (2009) (arguing that under an institutional analysis, as opposed to a pure legal process analysis, asbestos litigation serves as a reasonable use of judicial power).

      [15].   Ian Ayres & John Braithwaite, Tripartism: Regulatory Capture and Empowerment, 16 Law & Soc. Inquiry 435, 451–56 (1991).

      [16].   Id.

      [17].   Except in narrow circumstances, the FAA would not waive sanctions for problems resulting from criminal behavior or for problems disclosed in anticipation of an FAA inspection or during one.  Mills, supra note 13, at 17–19.  Other agencies have also adopted voluntary reporting programs.  See, e.g., Jodi L. Short & Michael W. Toffel, Coerced Confessions: Self-Policing in the Shadow of the Regulator, 24 J.L. Econ. & Org. 45, 46, 62–65 (2008).

      [18].   Russell W. Mills, The Development of Collaborative Regulatory Partnerships with Industry: A Historical Institutionalist Investigation of the Federal Aviation Administration’s Voluntary Safety Reporting Programs 26 (Apr. 2, 2010) (unpublished manuscript) (on file with author) [hereinafter Mills, Collaborative Regulatory Partnerships]; see also Mills, Collaborative Voluntary Partnerships, supra note 13.

      [19].   14 C.F.R. § 193 (2011).

      [20].   Id.; see also 5 U.S.C. § 552 (2006).

      [21].   Mills, supra note 13, at 28.

      [22].   Airworthiness directives are rules through which the FAA requires airlines to employ certain safety measures in certain types of aircraft.  See id. at 12.

      [23].   Press Release, Fed. Aviation Admin., FAA Proposes $10.2 Million Civil Penalty Against Southwest Airlines (Mar. 6, 2008), available at; Judson Berger, Southwest Has History of Triggering FAA Action, (Apr. 4, 2011),

      [24].   Mills, Collaborative Regulatory Partnerships, supra note 18, at 4–5.

      [25].   Id. at 5–6.

      [26].   The relevant part broke up and immediately lost all functionality.  Walter Berry & Lien Hoang, ‘Fuselage Rupture’ Forces Emergency Landing On Southwest Airlines Flight, Huffington Post (Apr. 2, 2011),

-landing-southwest_n_843925.html.  The plane shed five feet of exterior skin, “an approximately 9-inch wide by 59-inch long rectangular-shaped hole,” and opened the passenger cabin to the outside environment, which was the stratosphere.  Rapid Decompression Due to Fuselage Rupture, Nat’l Transp. Safety Board, (last visited Sept. 2, 2012).  That in itself is life threatening even if the plane does not fall out of the sky.  See id.

      [27].   Press Release, Fed. Aviation Admin., FAA Will Mandate Inspections for Early Models of 737 Aircraft (Apr. 4, 2011), available at

      [28].   The NTSB’s press release on the accident suggested that all maintenance inspections were conducted, there were no discrepancies, and there was no sign of apparent wrongdoing.  Press Release, Nat’l Transp. Safety Bd., NTSB Continues Investigation of Southwest Flight 812 (Apr. 25, 2011), available at  What seemed to happen is that the FAA, the airline, and the manufacturer did not believe the specific part of the plane warranted attention on a plane of this age; therefore, they did not expect cracks in it, and so no one checked for it.  NTSBgov, NTSB Final Press Briefing SWA Flight 812 Apr 4 2011, YouTube (Apr. 4, 2011),  “It was not an area that was believed could fail,” said the NTSB board member.  Id.

      [29].   FAA Docket No. 2011-0348, Airworthiness Directive 7 (May 6, 2011), available at$FILE/2011-08-51.pdf.

      [30].   Mills, supra note 13, at 32–33.  This view is also raised in the medical context.  See, e.g., Michael R. Cohen, Why Error Reporting Systems Should be Voluntary, 320 Brit. Med. J. 728, 729 (2000).  Cohen, president of the Institute for Safe Medication Practices and a well-credentialed pharmacist and scholar, argues that when disclosure is used primarily to punish the wrongdoer (creating a “fear of retribution”) rather than to create solutions, it has a stifling effect on disclosure: “[N]on-punitive and confidential voluntary reporting programmes provide more useful information about errors and their causes than mandatory reporting programmes. . . .  Practitioners who are forced to report errors are less likely to provide in depth information because their primary motivation is self protection and adherence to a requirement . . . .”  Id. at 729.

      [31].   Mills, supra note 13, at 17.

      [32].   U.S. Gov’t Accountability Office, GAO-10-414, Aviation Safety: Improved Data Quality and Analysis Capabilities are Needed as FAA Plans a Risk-Based Approach to Safety Oversight (2010), available at

      [33].   Russell W. Mills, Collaborating with Industry to Ensure Regulatory Oversight: The Use of Voluntary Safety Reporting Programs by the Federal Aviation Administration 156–57 (May 2011) (unpublished Ph.D. dissertation, Kent State University), available at

      [34].   Edward W. Stimpson et al., Managing Risks in Civil Aviation: A Review of the FAA’s Approach to Safety 56 (2008), available at

      [35].   Id. at 65 app. 3.

      [36].   Following a reorganization in 2010, the agency was restructured and is now named the “Bureau of Ocean Energy Management, Regulation and Enforcement.”  Salazar Swears-In Michael R. Bromwich to Lead Bureau of Energy Management, Regulation and Enforcement, Bureau of Ocean Energy Mgmt., Reg. & Enforcement (Jun. 21, 2010),  Here, however, all the examples described date from before the reorganization, and the sources I used refer to the agency as the “Mineral Management Service.”  I therefore considered it less confusing to use the old name here.

      [37].   Memorandum from Earl E. Devaney, Inspector Gen., U.S. Dep’t of the Interior, to Dirk Kempthorne, Sec. of the Interior 1–3 (Sept. 9, 2008), available at
-OIG_Report_2008.pdf (regarding OIG investigations of MMS employees).

      [38].   Dep’t of the Interior, Office of Inspector Gen., Investigative Report of MMS Oil Marketing Group – Lakewood 1 (2008) [hereinafter Investigative Report],available at .

      [39].   Id. at 1–2.

      [40].   Id. at 1–2, 4.

      [41].   Id. at 10–31.

      [42].   Id. at 11.

      [43].   See, e.g., Perry Bacon, Jr. et al., Lawmakers Assail Minerals Management Service, Wash. Post (May 26, 2010), (explaining that members of both parties of Congress criticized the MMS); see also Kathy Finn, Two Years After BP Oil Spill, Tourists Back in U.S. Gulf, Reuters (May 27, 2012), (discussing oil spill from BP oil well in Gulf of Mexico on April 20, 2010).

      [44].   The Deepwater Horizon Incident: Are the Minerals Management Service Regulations Doing the Job? Hearing Before the S. Comm. on Energy & Mineral Res., 111th Cong. 13–15 (2010) (statement of Mary L. Kendall, Acting Inspector General, Department of the Interior), available at

      [45].   Jaclyn Lopez, BP’s Well Evaded Environmental Review: Categorical Exclusion Policy Remains Unchanged, 37 Ecology L. Currents 93, 95–98 (2010).

      [46].   But see Christopher Carrigan, Minerals Management Service and Deepwater Horizon: What Role Should Capture Play?, in Preventing Capture: Special Interest Influence in Regulation, and How to Limit It (Daniel Carpenter & David Moss eds., forthcoming) (manuscript at 56–58) (discussing that the MMS was not, as a whole, captured).

      [47].   Bernstein, supra note 3, at 268; Jean-Jacques Laffont & Jean Tirole, A Theory of Incentives in Procurement and Regulation 475–80 (1993); Jon Hanson & David Yosifon, The Situation: An Introduction to the Situational Character, Critical Realism, Power Economics, and Deep Capture, 152 U. Pa. L. Rev. 129, 202–05 (2003); Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1667, 1685 (1975).

      [48].   Craig W. Thomas et al., Special Interest Capture of Regulatory Agencies: A Ten-Year Analysis of Voting Behavior on Regional Fishery Management Councils, 38 Pol’y Stud. J. 447, 448 (2010).  There is room in the literature for an article identifying different types of behavior that fall under the definition of capture and then addressing each type separately.  That is not this Article.  As the literature review demonstrates here, scholarship focusing on a specific type of capture, such as the revolving door, exists.  See, e.g., Ernesto Dal Bó, Regulatory Capture: A Review, 22 Oxford Rev. of Econ. Pol’y 203, 214–15 (2006).  But much of the literature speaks in much more general terms.  See, e.g., Steven P. Croley, Regulation and Public Interests: The Possibility of Good Regulatory Government 26–29 (2008); Malcolm K. Sparrow, The Regulatory Craft: Controlling Risks, Solving Problems, and Managing Compliance 35, 37 (2000); Ayres & Braithwaite,supra note 15, at 436–39; Makkai & Braithwaite, supra note 1.

      [49].   Paul J. Quirk, Industry Influence in Federal Regulatory Agencies 4 (1981).

      [50].   Daniel Carpenter & David Moss, Introduction, in Preventing Capture: Special Interest Influence in Regulation, and How to Limit It, supra note 46 (manuscript at 20).

      [51].   Makkai & Braithwaite, supra note 1, at 64, 66.

      [52].   James Kwak, Cultural Capture and the Financial Crisis, in Preventing Capture: Special Interest Influence in Regulation, and How to Limit It, supra note 46 (manuscript at 4–7).

      [53].   Susan Webb Yackee, Reconsidering Agency Capture During Regulatory Policymaking, in Preventing Capture: Special Interest Influence in Regulation, and How to Limit It, supra note 46 (manuscript at 10).

      [54].   See Ian Ayres & John Braithwaite, Responsive Regulation: Transcending the Deregulation Debate 55–56 (Donald R. Harris et al. eds., 1992) (“Where relationships are ongoing, where encounters are regularly repeated with the same regulator, corruption is more rewarding for both parties: the regulator can collect recurring bribe payments and the firm can benefit from repeated purchases of lower standards.  Moreover, ongoing relationships permit the slow sounding out of the corruptibility and trustworthiness of the other to stand by corrupt bargains (and at minimum risk because an identical small number of players are involved each time).”); Neil Gunningham, Assessing Responsive Regulation ‘on the Ground’: Where Does It Work? 3 (2011) (unpublished manuscript) (on file with the author).

      [55].   Letter from Richard Olney, Att’y Gen., to Charles E. Perkins, President of the Chi., Burlington & Quincy R.R. (1892), quoted in Louis L. Jaffe, The Effective Limits of the Administrative Process: A Reevaluation, 67 Harv. L. Rev. 1105, 1109 n.7 (1954).

      [56].   Bernstein, supra note 3, at 3–8.

      [57].   Id. at 83–102.

      [58].   For a summary of previous criticisms and an additional one, see Paul Sabatier, Social Movements and Regulatory Agencies: Toward a More Adequate—and Less Pessimistic—Theory of “Clientele Capture”, 6 Pol’y Sci. 301, 303–05 (1975).  Among the criticisms mentioned is that often legislation is designed to promote close connections between industry and regulator and that Bernstein’s assumptions that the agency cannot mobilize a supporting constituency to counteract industry influence and that the supporting coalition inevitably disperses or loses interest are problematic.  Id. at 305.

      [59].   See, e.g., Nader & Smith, supra note 11, at 59–60, 83–86; Turner, supra note 8, at 37–41, 120–21.

      [60].   William D. Berry, An Alternative to the Capture Theory of Regulation: The Case of State Public Utility Commissions, 28 Am. J. Pol. Sci. 524, 524–27 (1984); Sabatier, supra note 58, at 304.

      [61].   Croley, supra note 48, at 50–51; Richard A. Posner, The Behavior of Administrative Agencies, 1 J. Legal Stud. 305, 315–16 (1972).

      [62].   See generally George J. Stigler, The Theory of Economic Regulation, 2 Bell J. Econ. & Mgmt. Sci. 3 (1971) [hereinafter Stigler, Economic Regulation].  This essay followed a previous article by Stigler and Friedland, in which they examined the effect of regulation in the electricity sector and concluded it had little effect.  See George J. Stigler & Claire Friedland, What Can Regulators Regulate? The Case of Electricity, 5 J.L. & Econ. 1, 11–12 (1962).

      [63].   Stigler, Economic Regulation, supra note 62, at 3–7.

      [64].   Sam Peltzman, Toward a More General Theory of Regulation, 19 J.L. & Econ. 211, 211–13 (1976).  Among other things, Peltzman suggests that where industry influence conflicts with other influences, the result will usually not be completely caving to industry or to consumers but somewhere in between—though it will lean towards industry.  Id. at 227–30.

      [65].   Sam Peltzman, The Economic Theory of Regulation After a Decade of Deregulation, 1989 Brookings Papers on Econ. Activity Microeconomics 1, 4–5.

      [66].   Id.; see also Dal Bó, supra note 48, at 206–07.

      [67].   Dal Bó, supra note 48, at 204, 207–11.

      [68].   Croley, supra note 48, at 9–13, 20–21, 59–60; Fiona Haines, The Paradox of Regulation: What Regulation Can Achieve and What It Cannot 12–30 (2011); Dal Bó, supra note 48, at 204, 206.

      [69].   Dal Bó, supra note 48, at 215–16.

      [70].   Quirk, supra note 49, at 16–21.

      [71].   Dal Bó, supra note 48, at 207–13.

      [72].   Id. at 214–15.  Dal Bó also addresses two studies that suggested some positive influences to the revolving door.  Id. at 217–19.

      [73].   Cornelius M. Kerwin, Rulemaking: How Government Agencies Write Law and Make Policy 182–84 (3d ed. 2003) (“Businesses . . . are involved in rulemaking more often than are other groups, and they devote to it greater slices of their likely larger budgets and staffs.”); Berry, supra note 60, at 525–26; Croley,supra note 3, at 126–31; Marissa Martino Golden, Interest Groups in the Rule-Making Process: Who Participates? Whose Voices Get Heard?, 8 J. Pub. Admin. Res. & Theory 245, 259–64 (1998); Jason Webb Yackee & Susan Webb Yackee, A Bias Towards Business? Assessing Interest Group Influence on the U.S. Bureaucracy, 68 J. Pol. 128, 133–35 (2006).  But see Mariano-Florentino Cuéllar, Rethinking Regulatory Democracy, 57 Admin. L. Rev. 411, 435–59, 497–99 (2005) (finding substantial participation by lay people and public interest groups in three rulemakings and finding that the sophistication of the comments, rather than the identity of the commentator, determines whether a comment influences the agency’s decision).

      [74].   Shapiro & Steinzor, supra note 3, at 1752–55.

      [75].   Daniel P. Carpenter, Protection Without Capture: Product Approval by a Politically Responsive, Learning Regulator, 98 Am. Pol. Sci. Rev. 613, 614–15, 625–26 (2004).

      [76].   Carpenter & Moss, supra note 50 (manuscript at 10–13, 17) (elaborating upon and discussing the idea that claims of capture are often supported by sketchy and problematic evidence).

      [77].   In discussing the idea of “tripartism,” the authors suggest NGOs can serve as a counter to industry influence.  Ayres & Braithwaite, supra note 54, at 54–73.

      [78].   Id. at 65–69; Ayres & Braithwaite, supra note 15, at 452–56.

      [79].   Ayres & Braithwaite, supra note 54, at 65.

      [80].   Id.

      [81].   Makkai & Braithwaite, supra note 1, at 62–64.

      [82].   Id. at 64–66.

      [83].   Id. at 69–72.

      [84].   See, e.g., Deborah S. Dalton, Negotiated Rulemaking Changes EPA Culture, in Federal Administrative Dispute Resolution Deskbook 135, 135–52 (Marshall J. Breger et al. eds., 2001); Daniel J. Fiorino & Chris Kirtz, Breaking Down Walls: Negotiated Rulemaking at EPA, 4 Temp. Envtl. L. & Tech. J. 29, 29–30, 40 (1985); Daniel J. Fiorino, Regulatory Negotiation as a Form of Public Participation, in Fairness and Competence in Citizen Participation 223–25 (Ortwin Renn et al. eds., 1995); Daniel J. Fiorino, Regulatory Policy and the Consensus Trap: An Agency Perspective, 19 Analyse & Kritik 64, 74–75 (1997); Jody Freeman,Collaborative Governance in the Administrative State, 45 UCLA L. Rev. 1, 97–98 (1997); Philip J. Harter, Assessing the Assessors: The Actual Performance of Negotiated Rulemaking, 9 N.Y.U. Envtl. L.J. 32, 32–38, 52–54 (2000); Laura I. Langbein & Cornelius M. Kerwin, Regulatory Negotiation Versus Conventional Rule Making: Claims, Counterclaims, and Empirical Evidence, 10 J. Pub. Admin. Res. & Theory 599, 599–600 (2000); Siobhan Mee, Negotiated Rulemaking and Combined Sewer Overflows (CSOs): Consensus Saves Ossification?, 25 B.C. Envtl. Aff. L. Rev. 213, 213 (1997).

      [85].   Bardach & Kagan, supra note 12, at 134–40.

      [86].   Ayres & Braithwaite, supra note 54, at 25–35.  The authors expressly address the tension between their approach and capture, and suggest a solution, which I will address in Part I.C.

      [87].   Virginia Haufler, A Public Role for the Private Sector 4 (2001); Neil Gunningham & Joseph Rees, Industry Self-Regulation: An Institutional Perspective, 19 Law & Pol’y 363, 363–66 (1997).  For a detailed analysis of self-regulation, including its strengths and weaknesses, see Ayres & Braithwaite, supra note 54, at 102–28.

      [88].   For a recent overview of that literature, see Dominique Custos & John Reitz, Public-Private Partnerships, 58 Am. J. Comp. L. 555, 555 (2010).  For an article expressing concerns about the effects of public-private partnerships on accountability, see Martha Minow, Public and Private Partnerships: Accounting for the New Religion, 116 Harv L. Rev. 1229, 1255–59 (2003).  For examples of promoting such partnership in practice, see Noel P. Greis & Monica L. Nogueira, Food Safety—Emerging Public-Private Approaches: A Perspective for Local, State, and Federal Government Leaders 6–9 (2010); Cassandra Moseley, Strategies for Supporting Frontline Collaboration: Lessons from Stewardship Contracting 6–8 (2010).

      [89].   Kenneth M. Murchison, The Snail Darter Case: TVA Versus the Endangered Species Act 1–6, 193–98 (2007).

      [90].   George F. Wilhere, Three Paradoxes of Habitat Conservation Plans, 44 Envtl. Mgmt. 1089, 1089–90 (2009).

      [91].   Id. at 1090.

      [92].   Id. at 1089.

      [93].   Errol E. Meidinger, Laws and Institutions in Cross-Boundary Stewardship, in Stewardship Across Boundaries 87, 101 (Richard L. Knight & Peter B. Landres eds., 1998).

      [94].   AIBS Co-Sponsors HCPs Study, 48 BioScience 228, 228–29 (1998); Cameron W. Barrows et al., A Framework for Monitoring Multiple-Species Conservation Plans, 69 J. Wildlife Mgmt. 1333, 1343–44 (2005); Frances C. James, Lessons Learned from a Study of Habitat Conservation Planning, 49 BioScience 871, 873–74 (1999); Jennifer Jester, Habitat Conservation Plans Under Section 10 of the Endangered Species Act: The Alabama Beach Mouse and the Unfulfilled Mandate of Species Recovery, 26 B.C. Envtl. Aff. L. Rev. 131, 147–54 (1998); M. Nils Peterson et al., A Tale of Two Species: Habitat Conservation Plans as Bounded Conflict, 68 J. Wildlife Mgmt. 743, 743–45 (2004); Matthew E. Rahn et al., Species Coverage in Multispecies Habitat Conservation Plans: Where’s the Science?, 56 BioScience 613, 616–19 (2006); Wilhere, supra note 90, at 1089.

      [95].   Leigh Raymond, Cooperation Without Trust: Overcoming Collective Action Barriers to Endangered Species Protection, 34 Pol’y Stud. J. 37, 52–54 (2006); Craig W. Thomas, Habitat Conservation Planning: Certainly Empowered, Somewhat Deliberative, Questionably Democratic, 29 Pol. & Soc’y 105, 118–24 (2001).

      [96].   Lisa Blomgren Bingham, Avoiding Negotiation: Strategy and Practice, in The Negotiator’s Fieldbook 113, 113–20 (Andrea Kupfer Schneider & Christopher Honeyman eds., 2006); Amy J. Cohen, Negotiation, Meet New Governance: Interests, Skills, and Selves, 33 Law & Soc. Inquiry 503, 534 (2008).  For criticism of this development, see Jonathan D. Mester, The Administrative Dispute Resolution Act of 1996: Will the New Era of ADR in Federal Administrative Agencies Occur at the Expense of Public Accountability?, 13 Ohio St. J. on Disp. Resol. 167, 168–69 (1997).

      [97].   Martina Vidovic & Neha Khanna, Can Voluntary Pollution Prevention Programs Fulfill Their Promises?  Further Evidence from the EPA’s 33/50 Program, 53 J. Envtl. Econ. & Mgmt. 180, 180–82, 189–92 (2007).

      [98].   Mills, Collaborative Regulatory Partnerships, supra note 18, at 45.

      [99].   Ayres & Braithwaite,  supra note 54, at 55.

    [100].   Neal Shover et al., Enforcement or Negotiation: Constructing a Regulatory Bureaucracy 5 (1986); see also Barry M. Mitnick, The Political Economy of Regulation: Creating, Designing, and Removing Regulatory Forms 38 (1980); Sparrow, supra note 48, at 18, 35.

    [101].   Mark Seidenfeld, Empowering Stakeholders: Limits on Collaboration as the Basis for Flexible Regulation, 41 Wm. & Mary L. Rev. 411, 412 (2000).

    [102].   See, e.g., Ayres & Braithwaite, supra note 54, at 57–60; Moseley, supra note 88, at 9.Busenberg;Freeman;Chris Ansell & Alison Gash, Collaborative Governance in Theory and Practice, 18 Journal of Public Administration Research and Theory 543,  (2008);Archon Fung, et al., Full Disclosure: The Perils and Promise of Transparency (Cambridge University Press. 2007);Doug  Henton, et al., Collaborative Governance: A Guide for Grantmakers   (William and Flora Hewlett Foundation. 2006);Charles Koch, Collaborative Governance in the Restructured Electricity Industry, 40 Wake Forest L. Rev. 589,  (2005);Janet   Newman, et al., Public Participation and Collaborative Governance, 33 Journal of Social Policy 203,  (2004).

    [103].   Reuel E. Schiller, Enlarging the Administrative Polity: Administrative Law and the Changing Definition of Pluralism, 1945–1970, 53 Vand. L. Rev. 1389, 1405–13 (2000).

    [104].   Ayres & Braithwaite, supra note 54, at 57–60.

    [105].   Berry, supra note 60, at 525–26; Paul A. Sabatier et al., Hierarchical Controls, Professional Norms, Local Constituencies, and Budget Maximization: An Analysis of U.S. Forest Service Planning Decisions, 39 Am. J. Pol. Sci. 204, 221, 226, 229 (1995); Sabatier, supra note 58, at 325–26.

    [106].   Berry, supra note 60, at 542; William T. Gormley, Jr., Alternative Models of the Regulatory Process: Public Utility Regulation in the States, 35 W. Pol. Q. 297, 302–05, 309–10 (1982); Shapiro & Steinzor, supra note 3, at 1742; Guy L. F. Holburn & Pablo T. Spiller, Interest Group Representation in Administrative Institutions: The Impact of Consumer Advocates and Elected Commissioners on Regulatory Policy in the United States 14–16 (Univ. of Cal. Energy Inst., Energy Policy & Econ. Working Paper No. 002, 2002).

    [107].   Peltzman points out that capture does not mean that the regulator will only follow the industry’s wishes; if there is a competing group, it might have some influence but less than that of the capturing group.  Peltzman, supra note 65, at 9–14; Peltzman, supra note 64, at 217–19.

    [108].   See, e.g., Sabatier et al., supra note 105, at 229–32.  The authors demonstrate that there is strong influence of community and environmental groups, and, in some cases, an influence equal to that of business groups.  Id.

    [109].   This is because some aspects of responsive regulations may not work in every case.  See Gunningham, supra note 54, at 1–4, 10–12.

    [110].   Sabatier, supra note 58, at 325–27.  The author presents this as a solution that is sometimes available, but certainly one that is not always available.  Id.

    [111].   Ayres & Braithwaite, supra note 15, at 444–45.

    [112].   Id. at 444.

    [113].   The argument is that, for many regulatory issues, affected members of the public have too low of a stake in the outcome to invest sufficiently in acquiring information and monitoring.  Peltzman, supra note 64, at 212–13.

    [114].   See Ayres and Braithwaite, supra note 15, at 472–73, 478–87.

    [115].   Croley, supra note 48, at 138–39; Jeffrey S. Lubbers, The Transformation of the U.S. Rulemaking Process—For Better or Worse, 34 Ohio N.U. L. Rev. 469, 471, 481–82 (2008); Dorit Rubinstein Reiss, Tailored Participation: Modernizing the APA Rulemaking Procedures, 12 N.Y.U. J. Legis. & Pub. Pol’y 321, 330–37 (2009).

    [116].   See Golden, supra note 73, at 247–48, 259–61 (noting that those with money, particularly businesses, are the most influential); Reiss, supra note 115, at 332 (noting that it is rare for others beyond interest groups, particularly business interest groups, to participate in rulemaking); Shapiro & Steinzor, supra note 3, at 1752–55 (noting that “many more business groups lobby the Executive Branch than public interest groups” (citaton omitted)); Webb Yackee & Webb Yackee, supranote 73, at 128–30 (“[B]usiness interests dominate bureaucratic policymaking . . . .”); William F. West, Formal Procedures, Informal Processes, Accountability, and Responsiveness in Bureaucratic Policy Making: An Institutional Policy Analysis, 64 Pub. Admin. Rev. 66, 66–68 (2004) (noting that public comment “was numerically weighted in favor of business groups”).

    [117].   Bernstein, supra note 3, at 269–71.

    [118].   Id.; Sabatier, supra note 58, at 317–20.

    [119].   Ayres & Braithwaite, supra note 15, at 439–40.

    [120].   Id. at 440.

    [121].   Anne Bishop, Daughter of the Blood 108 (1998).

    [122].   Bernstein, supra note 3, at 270–71; Ayres & Braithwaite, supra note 15, at 449; Hanson & Yosifon, supra note 47, at 214 (explaining that the second layer of capture is where “the interior situation of relevant actors is also subject to capture. . . . [T]argets include the way that people think and the way that they think they think.”).

    [123].   Stigler & Friedland, supra note 62.

    [124].   Amitai Etzioni, Does Regulation Reduce Electricity Rates? A Research Note, 19 Pol’y Sci. 349, 351–52 (1986).

    [125].   Carpenter, supra note 75, at 614.

    [126].   Id. at 614–15.

    [127].   Cf. Robert D. Behn, Rethinking Democratic Accountability 3 (2001) (“Those whom we want to hold accountable have a clear understanding of what accountability means: Accountability means punishment. . . .  Moreover, the definition of a ‘screwup’ is constantly changing. . . .  Public officials may not realize that something is a ‘screwup’ until someone holds them accountable for doing what many others have been doing for quite a while.”).

    [128].   Croley, supra note 48, at 28–29; Peltzman, supra note 65, at 9–10; Paul Eric Teske et al., The Economic Theory of Regulation and Trucking Deregulation: Shifting to the State Level, 79 Pub. Choice 247, 248–49 (1994).

    [129].   See Croley, supra note 48, at 14; Cary Coglianese & David Lazer, Management-Based Regulation: Prescribing Private Management to Achieve Public Goals, 37 Law & Soc’y. Rev. 691, 696, 698, 700 (2003); Cary Coglianese et al., Performance-Based Regulation: Prospects and Limitations in Health, Safety, and Environmental Protection, 55 Admin. L. Rev. 705, 706, 711–15, 723 (2003); Lars Noah, The Little Agency That Could (Act with Indifference to Constitutional and Statutory Strictures), 93 Cornell L. Rev. 901, 901–04 (2008).

    [130].   Robert Sprague & Aaron J. Lyttle, Shareholder Primacy and the Business Judgment Rule: Arguments for Expanded Corporate Democracy, 16 Stan. J.L. Bus. & Fin. 1, 4–5 (2010).  While still dominant in practice, that notion is challenged in some corporate law scholarship.  See, e.g., Robert Charles Clark, Corporate Law 677–81 (1986).  The author claims that corporations have duties to society at large, not just to their shareholders, and that those duties require corporate responsibility.  Id. at 695; see also C. James Koch, Social Responsibility, Corporate Strategy and Profits, 1 Harv. Envtl. L. Rev. 662, 664–68 (1976); Colin Scott,Reflexive Governance, Meta-Regulation and Corporate Social Responsibility: The ‘Heineken Effect’, in Perspectives on Corporate Social Responsibility 178–82 (Nina Boeger et al. eds., 2008).

    [131].   Dion Casey, Agency Capture: The USDA’s Struggle to Pass Food Safety Regulations, Kan. J.L. & Pub. Pol’y, Spring 2008, at 142, 147–48.

    [132].   Id.

    [133].   Id.

    [134].   Id. at 153–56.

    [135].   Etzioni, supra note 3, at 320.

    [136].   Id.

    [137].   Id.

    [138].   Id.

    [139].   Id. at 321.

    [140].   Id.

    [141].   Daniel Carpenter, Corrosive Capture? The Dueling Forces of Autonomy and Industry Influence in FDA Pharmaceutical Regulation, in Preventing Capture: Special Interest Influence in Regulation, and How to Limit It, supra note 46 (manuscript at 3–5).

    [142].   Peter Grabosky & John Braithwaite, Of Manners Gentle: Enforcement Strategies of Australian Business Regulatory Agencies 198–201, 203–07 (1987); Christopher D. Stone, Where the Law Ends: The Social Control of Corporate Behavior 93–110 (1st ed. 1975); Turner, supra note 8, at 17–18, 37–45; Etzioni, supranote 3, at 320.

    [143].   Sparrow, supra note 48, at 35; Dal Bó, supra note 48, at 214–15, 217–18.  But for a more positive view of the revolving door, see Ayres & Braithwaite,supra note 15, at 436–37.

    [144].   Memorandum from John E. Dupuy, Assistant Inspector Gen. for Investigations, to S. Elizabeth Birnbaum, Dir., Minerals Mgmt. Serv. (Apr. 12, 2010) (on file with the author).

    [145].   Memorandum from Mary L. Kendall, Acting Inspector Gen., to Ken Salazar, Sec. of the Interior (May 24, 2010) (on file with the author).  But seegenerally Carrigan, supra note 46 (suggesting a different, historical explanation from Kendall’s assessment of the problem and its source).

    [146].   Nader & Smith, supra note 11, at 99–101.

    [147].   42 U.S.C. § 15942 (2006).

    [148].   U.S. Gov’t Accountability Office, GAO-09-872, Energy Policy Act of 2005: Greater Clarity Needed to Address Concerns with Categorical Exclusions for Oil and Gas Development Under Section 390 of the Act 12 (Sept. 2009), available at

    [149].   Id. at 23–29.  Note that the GAO did not attribute these deviations to wrongdoing: “We did not find intentional actions on the part of BLM staff to circumvent the law; rather, our findings reflect what appear to be honest mistakes stemming from confusion in implementing a new law with evolving guidance.” 29; see also Juliet Eilperin, U.S. Exempted BP Rigs from Impact Analysis, Wash. Post, May 5, 2010, at A4.

    [150].   42 U.S.C. § 4332 (2006).

    [151].   Eilperin, supra note 149.

    [152].   Id. (quoting Kierán Suckling, Exec. Dir., Ctr. for Biological Diversity).

    [153].   Haines, supra note 68, at 50–52, 203; David Levi-Faur, Regulatory Capitalism: The Dynamics of Change Beyond Telecoms and Electricity, 19 Governance 497, 503–04 (2006).

    [154].   See Carpenter, supra note 75, at 613 (focusing on the entry-barrier aspect of capture).

    [155].   Damien Geradin, The Opening of State Monopolies to Competition: Main Issues of the Liberalization Process, in The Liberalization of State Monopolies in the European Union and Beyond 181, 182–84 (Damien Geradin ed., 2000); Pierre Larouche, Telecommunications, in The Liberalization of State Monopolies in the European Union and Beyond, supra, at 15, 44–45; Michael J. Legg, Verizon Communications, Inc. v. FCC—Telecommunications Access Pricing and Regulator Accountability Through Administrative Law and Takings Jurisprudence, 56 Fed. Comm. L.J. 563, 565–67, 575 (2004).

    [156].   Steven K. Vogel, Freer Markets, More Rules: Regulatory Reform in Advanced Industrial Countries 65–66, 88–92 (1996); Viktor Mayer-Schönberger & Mathias Strasser, A Closer Look at Telecom Deregulation: The European Advantage, 12 Harv. J.L. & Tech. 561, 564–66 (1999); Vincent Wright, Public Administration, Regulation, Deregulation and Reregulation, in Managing Public Organizations: Lessons from Contemporary European Experience 244, 245 (Kjell A. Eliassen & Jan Kooiman eds., 1993).

    [157].   In telecommunications, see the articles described in Dal Bó, supra note 48, at 216.  Beyond utilities, the railroads’ capture of the ICC led to increased fares.  See Samuel P. Huntington, The Marasmus of the ICC: The Commission, the Railroads, and the Public Interest, 61 Yale L.J. 467, 478, 480–81 (1952); Teske et al., supra note 128, at 249.

    [158].   Huntington, supra note 157, at 481–85.

    [159].   See supra note 66.

    [160].   Grabosky & Braithwaite, supra note 142, at 190–95; Keith Hawkins, Environment and Enforcement: Regulation and the Social Definition of Pollution 7 (1984); Ayres & Braithwaite, supra note 15, at 457–58 n.54; Seidenfeld, supra note 101, at 419, 424–25; Gunningham, supra note 54, at 4–8.

    [161].   Haines, supra note 68, at 24–26.

    [162].   Ayres & Braithwaite, supra note 15, at 453 (noting that the regulated firm is also a member of society, and increasing its welfare should count in calculating the effects to the general welfare).

    [163].   For a description of the APA framework and its goals, see Reiss, supra note 115, at 326; Reuel E. Schiller, Rulemaking’s Promise: Administrative Law and Legal Culture in the 1960s and 1970s, 53 Admin. L. Rev. 1139, 1140, 1145, 1162–63 (2001); Stewart, supra note 47, at 1713–14.  A thorough discussion of the APA is beyond this paper.

    [164].   5 U.S.C. § 553 (2006); see also Mathew D. McCubbins et al., Administrative Procedures as Instruments of Political Control, 3 J.L. Econ. & Org. 243, 258 (1987); Richard B. Stewart, Administrative Law in the Twenty-First Century, 78 N.Y.U. L. Rev. 437, 446–49 (2003); West, supra note 116, at 67.

    [165].   Lubbers, supra note 115, at 476–78.

    [166].   Sparrow, supra note 48, at 35; Richard J. Pierce, Jr., Judicial Review of Agency Actions in a Period of Diminishing Agency Resources, 49 Admin. L. Rev. 61, 62–64 (1997); see also Croley, supra note 48, at 17.

    [167].   Paul Pierson, From Expansion to Austerity: The New Politics of Taxing and Spending, in Seeking the Center: Politics and Policymaking at the New Century 54, 60–61, 73 (Martin A. Levin et al. eds., 2001).  The author uses the term “fiscal austerity” to discuss the possible retrenchment of the welfare state due to economic troubles of the twentieth century.  Id. at 55.  The term, however, also nicely reflects the current reality of financial crises and the need to tighten government spending—a reality felt at least for several decades.

    [168].   Shapiro & Steinzor, supra note 3, at 1758–62.

    [169].   Roger W. Cobb & David M. Primo, The Plane Truth: Airline Crashes, the Media, and Transportation Policy 16 (2003).

    [170].   Mills, supra note 13, at 12.

    [171].   Id.

    [172].   Casey, supra note 131, at 146.

    [173].   David C. Vladeck, The FDA and Deference Lost: A Self-Inflicted Wound or the Product of a Wounded Agency? A Response to Professor O’Reilly, 93 Cornell L. Rev. 981, 983–84 (2008); Lorna Zach & Vicki Bier, Risk-Based Regulation for Import Safety, in Import Safety: Regulatory Governance in the Global Economy 151, 153–54 (Cary Coglianese et al. eds., 2009).

    [174].   See Thomas O. Sargentich, The Critique of Active Judicial Review of Administrative Agencies: A Reevaluation, 49 Admin. L. Rev. 599, 602–03 (1997); David Schoenbrod, The EPA’s Faustian Bargain, Regulation, Fall 2006, at 36, 41–42; Vladeck, supra note 173, at 984–85 (arguing that, until Congress provides the FDA with adequate funding to protect public health, the agency will be unable to ameliorate its reputation).

    [175].   Pierce, supra note 166, at 62–65.

    [176].   See Sabatier et al., supra note 105, at 207 (“Given that information is costly to acquire and that individuals have limited information-processing capabilities, information must be condensed as it moves up the hierarchy.  Such condensation provides an opportunity for distortion, usually to flatter the officials involved and to mirror their policy views.  As a result, top officials may hold incomplete, and often biased views of the situations confronted by their subordinates.”).

    [177].   Alexander S. P. Pfaff & Chris William Sanchirico, Environmental Self-Auditing: Setting the Proper Incentives for Discovery and Correction of Environmental Harm, 16 J.L. Econ. & Org. 189, 189–91 (2000).

    [178].   Mills, supra note 13, at 14; Anna Alberini & Kathleen Segerson, Assessing Voluntary Programs to Improve Environmental Quality, 22 Envtl. & Resource Econ. 157, 158 (2002).

    [179].   Edward P. Weber & Anne M. Khademian, Wicked Problems, Knowledge Challenges, and Collaborative Capacity Builders in Network Settings, 68 Pub. Admin. Rev. 334, 343 (2008) (“[U]sing only government-based public managers and coercion to solicit information and bring about compliance may lead to short-term, incomplete, high-cost successes at the expense of long-term problem-solving effectiveness . . . .”).

    [180].   Dal Bó, supra note 48, at 214.

    [181].   Wagner, supra note 3, at 1331.

    [182].   David Shenk, Data Smog: Surviving the Information Glut 15–16 (1997); Wagner, supra note 3, at 1331.

    [183].   Archon Fung et al., Full Disclosure: The Perils and Promise of Transparency 171–72 (2007); Shenk, supra note 182; Ira S. Nathenson, Internet Infoglut and Invisible Ink: Spamdexing Search Engines with Meta Tags, 12 Harv. J.L. & Tech. 43, 51–53 (1998).

    [184].   Alberini & Segerson, supra note 178 (“[I]ncreased cooperation between polluters and regulators can improve information flows and reduce implementation lags.”).  On the importance of the format in which the information is presented (easy versus hard to digest), see Fung et al., supra note 183, at 57–64.

    [185].   Mary F. Evans et al., Regulation with Direct Benefits of Information Disclosure and Imperfect Monitoring, 57 J. Envtl. Econ. & Mgmt. 284, 285–86 (2009).

    [186].   A. Mitchell Polinsky & Steven Shavell, Mandatory Versus Voluntary Disclosure of Product Risks 4 (Nat’l Bureau of Econ. Research, Working Paper No. 12776, 2006), available at (“[V]oluntary disclosure will induce firms to acquire more information about product risks because they can keep silent if the information is unfavorable.”).

    [187].   Max H. Bazerman, Judgment in Managerial Decision Making 34–35 (5th ed. 2002).

    [188].   Id.

    [189].   Paul A. Sabatier, An Advocacy Coalition Framework of Policy Change and the Role of Policy-Oriented Learning Therein, 21 Pol’y Sci. 129, 133 (1988);see also William D. Leach & Paul A. Sabatier, To Trust an Adversary: Integrating Rational and Psychological Models of Collaborative Policymaking, 99 Am. Pol. Sci. Rev. 491, 494 (2005).

    [190].   Frédéric Boehm, Regulatory Capture Revisited – Lessons from Economics of Corruption 11, 16 (Internet Ctr. for Corruption Research, Working Paper No. 22, 2007), available at  Lying is also what the Nader report accuses the airlines of doing to the FAA.  See Nader & Smith, supra note 11, at 99–101.

    [191].   Bardach & Kagan, supra note 12, at 210–13 (noting the “regulatory ratchet” limitation on flexible regulatory enforcement and the persistence of unreasonableness).

    [192].   Nader & Smith, supra note 11, at 99–101.

    [193].   Turner, supra note 8, at 99–106.

    [194].   Grabosky & Braithwaite, supra note 142, at 190–91, 198–201; Roberto Pires, Promoting Sustainable Compliance: Styles of Labour Inspection and Compliance Outcomes in Brazil, 147 Int’l Labour Rev. 199, 200–02 (2008); John T. Scholz, Cooperation, Deterrence, and the Ecology of Regulatory Enforcement, 18 Law & Soc’y Rev. 179, 179–80, 185–87 (1984) [hereinafter Scholz, Cooperation]; Jodi L. Short & Michael W. Toffel, Making Self-Regulation More Than Merely Symbolic: The Critical Role of the Legal Environment, 55 Admin. Sci. Q. 361, 366–69 (2010).  But see generally John T. Scholz & Wayne B. Gray, Can Government Facilitate Cooperation? An Informational Model of OSHA Enforcement, 41 Am. J. Pol. Sci. 693 (1997) (emphasizing the importance of sanctions and supporting the view that they make a difference); John T. Scholz & Wayne B. Gray, OSHA Enforcement and Workplace Injuries: A Behavioral Approach to Risk Assessment, 3 J. Risk & Uncertainty 283 (1990) (emphasizing the importance of penalizing those who do not comply and supporting the view that penalties make a difference).

    [195].   Short & Toffel, supra note 194, at 368.

    [196].   Joseph V. Rees, Reforming the Workplace: A Study of Self-Regulation in Occupational Safety 12–13 (Keith Hawkins & John M. Thomas eds., 1988); Short & Toffel, supra note 194, at 368–69.  This mirrors the famous “bargaining in the shadow of the law” insight.  See Malcolm Feeley, Coercion and Compliance: A New Look at an Old Problem, in Compliance and the Law: A Multi-Disciplinary Approach 51, 60–62 (Samuel Krislov et al. eds., 1972); Robert H. Mnookin & Lewis Kornhauser, Bargaining in the Shadow of the Law: The Case of Divorce, 88 Yale L.J. 950, 950–51 (1979).

    [197].   Neal Shover, The Season of Responsive Regulation 7, 10–11 (June 3, 2011) (unpublished manuscript) (on file with the author).  For a discussion of the literature on the issue and many more citations, see id. at 6–13.

    [198].   Grabosky & Braithwaite, supra note 142, at 190–91, 203; Pires, supra note 194, at 200.

    [199].   Bardach & Kagan, supra note 12, at 292–97; Kagan, Adversarial Legalism, supra note 12, at 198, 200–04; Robert A. Kagan, The Consequences of Adversarial Legalism, in Regulatory Encounters: Multinational Corporations and American Adversarial Legalism 372, 373–74, 389–94, 400–05 (Robert A. Kagan & Lee Axelrad eds., 2000) [hereinafter Kagan, Consequences of Adversarial Legalism]; Pires, supra note 194; Weber & Khademian, supra note 179.

    [200].   Louis Kaplow & Steven Shavell, Optimal Law Enforcement with Self-Reporting of Behavior, 102 J. Pol. Econ. 583, 583–85 (1994); Michael W. Toffel & Jodi L. Short, Coming Clean and Cleaning Up: Does Voluntary Self-Reporting Indicate Effective Self-Policing?, 54 J.L. & Econ. 609, 618 (2011).

    [201].   For example, many enforcement actions by the EPA involve litigation.  See 42 U.S.C. §§ 7413, 9607 (2006); see also R. Shep Melnick, Regulation and the Courts: The Case of the Clean Air Act 200–02 (1983); Lynn Peterson, Promise of Mediated Settlements of Environmental Disputes: The Experience of EPA Region V, 17 Colum. J. Envtl. L. 327, 327–30 (1992).

    [202].   Kagan, Adversarial Legalism, supra note 12, at 29–32; James Q. Wilson, Bureaucracy: What Government Agencies Do and Why They Do It 282–84 (1989).

    [203].   Kagan, Consequences of Adversarial Legalism, supra note 199, at 380–81; Robert A. Kagan et al., Explaining Corporate Environmental Performance: How Does Regulation Matter?, 37 Law & Soc’y Rev. 51, 51–53, 82–84 (2003).

    [204].   Scholz, Cooperation, supra note 194, at 204, 208.

    [205].   Id. at 180–81.

    [206].   Bardach & Kagan, supra note 12, at 104–07.

    [207].   Id. at 64–66, 107–09.  This effect is termed “minimal compliance” by the authors.  Id. at 107–09.

    [208].   Id. at 109–11.

    [209].   Id. at 112–16.

    [210].   Edward L. Deci et al., A Meta-Analytic Review of Experiments Examining the Effects of Extrinsic Rewards on Intrinsic Motivation, 125 Psychol. Bull. 627, 627–28, 658–59 (1999); Deepak Malhotra & J. Keith Murnighan, The Effects of Contracts on Interpersonal Trust, 47 Admin. Sci. Q. 534, 534–35 (2002); Scholz,Cooperation, supra note 194, at 179–80; Short & Toffel, supra note 194, at 367–69.

    [211].   Grabosky & Braithwaite, supra note 142, at 203; Hawkins, supra note 160, at 3–6, 182–88; Makkai & Braithwaite, supra note 1, at 77.

    [212].   Mills, supra note 13, at 24.

    [213].   Haufler, supra note 87, at 3–4; Gunningham & Rees, supra note 87, at 401–02.

    [214].   Mills, supra note 13, at 25.

    [215].   Cass R. Sunstein, Risk and Reason: Safety, Law, and the Environment 144 (2002); Elizabeth Heger Boyle, Political Frames and Legal Activity: The Case of Nuclear Power in Four Countries, 32 Law & Soc’y Rev. 141, 155 (1998).

    [216].   Alicia Mundy, Generic Drug Makers Line Up Behind Proposal for FDA Fees, Wall St. J., Feb. 16, 2011, at B1.

    [217].   Alberini & Segerson, supra note 178, at 157–58; Vidovic & Khanna, supra note 97, at 182.

    [218].   Julian Le Grand, Knights, Knaves or Pawns? Human Behaviour and Social Policy, 26 J. Soc. Pol’y 149, 149, 154 (1997).

    [219].   Grabosky & Braithwaite, supra note 142, at 183–84; Gunningham, supra note 54, at 10–11.

    [220].   See Kagan, Adversarial Legalism, supra note 12, at 191–94.  Kagan demonstrates, citing extensive literature, that strict enforcement and adversarial approaches in the United States did not produce better regulatory results than in other countries and very likely resulted in the expenditure of more time and money to achieve essentially equivalent results.  Id. at 194–95, 198–205.

    [221].   Initially, this sentence read: “Nobody plans regulation with the intent to harm industry.”  But my colleague, John Leshy, pointed out that statement’s inaccuracy.  For example, he suggests, some mining regulation actually serves to restrict, and even undermine, mining operations.

    [222].   Bardach & Kagan, supra note 12, at 313–15; Clyde Wayne Crews, Jr., Ten Thousand Commandments: An Annual Snapshot of the Federal Regulatory State 1–2 (2008); Winston Harrington, Grading Estimates of the Benefits and Costs of Federal Regulation 1–4 (Res. for the Future, Discussion Paper No. 06-39, 2006), available at

    [223].   Geoffrey Garrett, Shrinking States? Globalization and National Autonomy in the OECD, 26 Oxford Dev. Studies 71, 95–96 (1998).

    [224].   Grabosky & Braithwaite, supra note 142, at 215–17; Cosmo Graham, Regulating Public Utilities: A Constitutional Approach 153–54 (2000).

    [225].   For a telling example, see Kagan’s description of the Oakland Port dredging and the effects the regulatory framework had on it.  Kagan, Adversarial Legalism, supra note 12, at 25–29.

    [226].   On the reasons for regulation, see Shapiro & Steinzor, supra note 3, at 1741.

    [227].   Short & Toffel, supra note 194, at 386–87 (suggesting that monitoring can increase compliance, though sanctions may not).  For a discussion of the literature on the issue and many more citations, see id. at 371.

    [228].   This may lead them to try to hide problems, but in a complex corporation, upper management may itself be unaware of the problems until very late, and finding the information sooner will be important to them.

    [229].   Makkai & Braithwaite, supra note 1, at 73.

    [230].   Cobb & Primo, supra note 169, at 17; see also Nader & Smith, supra note 11, at 61–68.

    [231].   Investigative Report, supra note 38, at 1–4.

    [232].   Komesar, supra note 14, at 6.

    [233].   Id.