In July 2021, a sweeping Executive Order committed the entire U.S. federal government to reining in big business. Dozens of proposed bills at the state level similarly target big business for stricter regulatory treatment. But unlike in past decades, today’s calls to break up and intensely regulate big business do not hinge on harms to consumers qua consumers. Instead, today’s anti-bigness sentiment rests to a large extent on the claim that big is bad because it is ungovernable. Giant corporations with market power treat legal requirements as mere recommendations, and routinely engage in behavior that harms our civil liberties and degrades the environment as long as it maximizes their own bottom line, or so the argument goes. But the “big is ungovernable” claim as currently construed is underdeveloped. In fact, many theoretical and empirical analyses suggest that big means better governability. If “big is ungovernable” is popular not because of the merits but strictly because of a strong anti-bigness sentiment, we could end up with bad policies negating economies of scale.
This Article provides a comprehensive assessment of how size and market power create governance problems, and a blueprint for how to ameliorate these problems. Super-big corporations have (1) power to shape the regulatory framework that governs them, and (2) a fragmentation of knowledge within them that makes it harder to detect and stop transgressions of the framework. These features dilute not just the expected legal sanction, but also moral constraints and the prospect of market discipline for misbehaving. For example, super-big corporations can leverage their market power to force their customers, workers, and suppliers to sign class action waivers and gag clauses, thereby diluting both legal and reputational sanctions for misbehaving. And the fragmentation of knowledge within super-big corporations makes it harder for law enforcers to detect culpability, and easier for individuals working within these corporations to behave badly without feeling guilty about it.
Recognizing the limitations of laws, markets, and morals to govern big business has concrete policy implications. At a general level, the analysis here can provide an intellectual structure for rethinking criteria for breaking up or downsizing firms according to governability. At a more specific level, this Article questions the desirability of oft-made proposals such as increasing the severity of sanctions. Solutions to bigness-control problems instead must directly address the power problem, such as by making class action waivers unenforceable, and the information problem, such as by recalibrating corporate law’s director oversight duties.
Introduction
For six decades, chemical giant DuPont discharged a highly toxic chemical dubbed C8 into the environment while making Teflon.1 The company’s decision makers realized by the 1980s that C8 is extremely toxic, that it does not break up in the environment and accumulates over time in the human body, and that it had leaked into the drinking water of nearby communities.2 Yet they refused seemingly inexpensive abatement and continued production for three more decades.3 Internal documents show that DuPont’s decision makers discussed in real time the possibility of suffering their “biggest tort issue” yet and reputational fallout if news of C8 got out.4 That they nevertheless continued polluting raises the possibility that they were counting on their ability to keep concealing information from the public and regulators, and to dilute the expected sanction. The problem is that they were right: a cost-benefit analysis suggests it was rational to pollute from a shareholder wealth maximization perspective.5
The DuPont-C8 case presents a challenge for conventional theories of corporate deterrence. DuPont is hardly a fly-by-night operator. It is rather one of the oldest and “most distinguished of . . . U.S. corporation[s],”6 enjoys a strong reputation for thinking long term and taking care of its stakeholders,7 and has the economies of scale to invest in abatement and compliance. In conventional wisdom, such well-established, reputation-sensitive companies are the ones least likely to engage in gross misconduct. Yet, the C8 case and a string of other epic-proportions debacles, such as Volkswagen’s emissions scandal and Wells Fargo’s phony-accounts scandal, suggest otherwise.8
Could it be that the biggest corporations are actually the least governable, unfazed by the prospects of legal liability and market discipline? This is hardly a theoretical question: “big is ungovernable” has reemerged as a rallying cry for today’s powerful anti-monopoly movement.9 Those who oppose breaking up or intensely regulating big business suggest that today’s monopolies are not the monopolies of old: they compete fiercely with each other and charge lower (sometimes zero) prices, thereby enhancing consumer welfare.10 Anti-monopolists rebut by pointing to the negative noneconomic effects of concentration: big is bad, they claim, because it undermines our civil liberties and democratic institutions. Big businesses’ disregard for the laws of society goes beyond illegal monopoly maintenance: it hurts our privacy, our environment, and the free press.11
Importantly, the “big is ungovernable” claim has by now firmly entered policy circles.12 In 2020, a congressional subcommittee investigating the conduct of big tech platforms maintained that the big platforms (Google, Amazon, Facebook, and Apple) leverage their power to shape the regulatory framework that governs them and repeatedly violate existing laws and court orders.13 This pattern of behavior, concluded the subcommittee, “raises questions about whether these firms view themselves as above the law, or whether they simply treat lawbreaking as a cost of business.”14 And in July 2021, President Biden signed a sweeping Executive Order containing seventy-two specific action items and ordering every part of the federal government (not just the antitrust enforcers) to focus on reining big business in.15
The stakes of the “big is ungovernable” claim could therefore not be higher. But the claim as currently construed is underdeveloped. In fact, there exist numerous theoretical and empirical analyses suggesting that big is more governable.16 Larger corporations are more publicly visible, increasing the probability that any misconduct would be detected. And they have more to lose from being caught misbehaving: from higher punitive damages in court to greater reputational fallout in the marketplace.17 If “big is ungovernable” is based on little more than anecdotes and riding a strong anti-bigness sentiment, we could end up with bad policies negating economies of scale.18
This, then, is the question at the heart of this Article: to what extent is big ungovernable? There are behaviors that society wishes to minimize, such as toxic pollution, worker-safety problems, and consumer fraud. And there are institutions meant to deter such misbehaviors, such as regulatory requirements, the threat of legal liability, reputation concerns, and anticipatory guilt feelings. This Article examines whether super-large firms with market power are less amenable to such legal and nonlegal controls.
This Article identifies several unique institutional features of big business that dilute the effectiveness of deterrence across all systems. Super-large firms can wield their power to influence the regulatory framework that governs them so that their behavior falls within the law. When these firms nevertheless transgress the lines, they can water down the prospect of both private and public enforcement. They leverage their market power to sign their counterparties on class action waivers and gag clauses, thereby diluting private enforcement. As for public enforcement, the fragmentation of knowledge inside these super-sized organizations makes it harder to prosecute top individual decision makers. And the firms’ systemic importance and big war chests make enforcers reluctant to vigorously pursue sanctions at the entity level, for fear of collateral consequences and the enforcers’ own reputation concerns. Bigness also dilutes the effectiveness of market discipline through increasing the costs for stakeholders of punishing the company by taking their business elsewhere, through blocking damning information from getting out, and through shaping how market actors interpret information that does get out. Finally, size also dilutes the moral constraints operating on individuals within large firms. Bigness increases the “moral wiggle room” that allows corporate actors to behave badly without feeling bad about it.19
Recognizing the gaps in the ability of laws, markets, and morals to govern big business opens up space for rethinking policy implications. This Article reveals why oft-made proposals to increase the sanctions and ramp up regulatory enforcement budgets are less likely to ameliorate bigness-control problems. Rather, solutions must directly address the two institutional features that render deterrence ineffective, namely, information and power asymmetries. To tackle information problems, we could promote whistleblowing by lower-level employees and emphasize the oversight duties of higher-level decision makers, in order to reverse their incentives to remain ignorant. To address problems stemming from power, we could make mandatory arbitration provisions unenforceable and recalibrate enforcers’ priorities. Enforcers should dedicate more of their scarce resources to vigorously pursuing meritorious cases against big business and to vigorously monitoring post-settlement behavior, even if this means opening fewer enforcement actions and collecting less money annually.
This Article’s contributions correspond to and weave together three timely topics, namely, antitrust, corporate purpose, and compliance. The revival of antitrust has coincided with calls to broaden the perspective beyond consumer welfare.20 This Article does just that, fleshing out the ways in which size and concentration affect firms’ propensities to engage in behaviors that may be good for the company’s bottom line but are inimical to broader societal interests. This, in turn, leads us to the corporate purpose debate. The analysis here explains why even Milton Friedman would admit that his maxim—the only social responsibility of business is to increase its profits21—does not apply to today’s corporate behemoths. Finally, compliance has become one of the most practically important topics, with trillions of dollars spent annually on compliance programs and trillions more lost due to the programs’ failures to prevent misconduct.22 Until recently, corporate law played a surprisingly limited role in compliance. This Article clarifies how recent doctrinal developments in director oversight duties can change things in ways that make corporate law a good antidote to bigness-control problems.
A few words on terminology and scope are in order from the outset. I define corporate misconduct for our purposes as behaviors within corporations that harm others or otherwise violate accepted legal and nonlegal rules.23 I do not focus on misconduct against the company, such as employee embezzlement, but rather on behaviors taken by representatives to benefit the company.24 Classic examples include polluting the environment and defrauding consumers.25 Within corporate misconduct, I highlight the differences between big and normal-sized corporations, where “big” denotes super-large corporations with market power.26 As this Article shows, while some of the control problems we discuss here increase monotonically with firm size, others appear only once the corporation crosses a certain bigness threshold.27 I therefore do not focus only on differences between a 100-employee company and a 1,000-employee one; the latter may be larger, but it is not super-large. Rather, I also focus on the differences between companies with 10,000 employees (think Chuck E. Cheese) and those with 100,000 (think Wells Fargo).28 In other words, bigness creates control problems that are not just greater in degree but also different in kind.
This Article proceeds in five parts. Part I situates the “big is ungovernable” claim against the backdrop of the broader anti-monopoly debate and offers an organizing framework to assess the claim. This Part synthesizes a multidisciplinary literature on corporate deterrence to highlight four general vectors that link bigness to governability. Super-large firms have (1) higher visibility; (2) more to lose from being caught misbehaving; (3) power to shape how their conduct is perceived; and (4) fragmentation of knowledge within them that makes detecting culpability extremely difficult. While the first two vectors push toward better governability, the latter two pull in the opposite direction. Answering which of the competing vectors dominate requires introducing context. This is where Parts II–IV come in, examining how each system of control (laws, reputation, and morals) deals with misconduct by massive corporations.
Part II breaks down the effectiveness of legal deterrence into two components. First, super-large corporations enjoy superior ability to influence the regulatory agenda, not just via lobbying or campaign contributions, but also via “soft” channels, such as epistemic capture or cultural capture. They can thus ensure that their behavior falls within the lines. Second, when their behavior nevertheless transgresses the lines, the unique institutional features of bigness hinder enforcers’ ability to investigate, prosecute, and properly calibrate the sanction for misbehavior. As a result, enforcers often avoid taking big business to court, opting instead to settle early for what big defendants write off as the small costs of doing business.
Part III evaluates how size affects market discipline (reputation concerns). Classic reputation models suggest that the effectiveness of reputational deterrence increases with firm size, as bigger companies have bigger reputational caches to protect. Part III finds this conventional wisdom wanting for various reasons. Market power makes it harder for consumers to switch to competitors, harder for workers to blow the whistle, and harder for information intermediaries to certify allegations. The biggest, most powerful firms in the industry often enjoy an aura of success, and market actors tend to perceive reports of misconduct by them as implausible or irrelevant. When Goldman Sachs does something shady, it may result in the firm getting credit for being the smartest in the room and pushing the envelope, whereas similar behavior by smaller actors would be condemned.
Part IV explains why size also dilutes the moral constraints operating on individuals within the firm. The fragmentation of knowledge and the hierarchical nature inside super-large corporations create moral blind spots. Looking from the outside in retrospect, we view behaviors such as emitting C8 as abhorrent. But inside mammoth-sized corporations in real time, no single individual realizes the moral consequences of her behavior. Size eases our natural tendencies toward selective perception, plausible deniability, and leaving ourselves “moral wiggle room.” It creates an environment where even good people are more likely to behave badly.
Part V discusses the implications of our review of the different ways in which bigness creates control problems. The Part starts by taking stock of the overall evidence and emphasizing the variation, by highlighting areas in which big business actually fares better. For example, big corporations seem to fare better (worse) on issues with high (low) saliency and low (high) complexity. When the lines are clear, they are better at staying within them. By contrast, when the lines are blurred, they are better at exploiting loopholes and pushing the envelope in ways that benefit them but harm society. Identifying these areas allows policymakers to focus on the more worrisome and fixable issues. Part V then emphasizes the need for policymakers to focus on the interactions between laws, markets, and morals. Even if each system is malleable, the combination of all three systems could provide a check on big business, as long as each system’s flaws are imperfectly correlated with the other systems’ flaws. Yet in reality, big corporations often use the flaws of one system to influence the other systems, such as using their clout over the regulator to also win in the court of public opinion.
Part V then continues to discuss concrete implications for regulators, judges, and academics. Regulators could employ the insights developed here to reevaluate the desirability of structural reforms versus conduct rules, or of recent concrete proposals such as the FAIR Act (meant to render mandatory arbitration provisions unenforceable). Judges could rely on the analysis here when interpreting shareholders’ right to inspect the company’s books and records (in corporate law), or when interpreting civil procedure doctrines on protective orders and confidential settlements. And academics could focus their attention on the connections that this Article exposes between seemingly disparate topics, such as regulatory capture and corporate governance or antitrust and corporate purpose.29 A short Conclusion juxtaposes this Article’s original contributions with the extant literature, recognizes this Article’s limitations, and offers big-picture observations on how to further develop bigness as a field of academic inquiry.
I. Background: The Emergence of the “Big is Ungovernable” Claim
In the heated debate over whether and how to regulate big business, anti-monopolists now frequently invoke the “big is ungovernable” claim. According to this claim, behavioral remedies, especially those that require continuous oversight, are useless for dealing with super-large corporations. Only structural solutions (“break them up!”) could safeguard users’ privacy, the political process’s integrity, the free press, and market dynamism, or so the argument goes. Section I.A explains how the “big is ungovernable” claim emerged. Section I.B provides an organizing framework to more rigorously think about the links between size and governability.
A. The Revival of Anti-Bigness Sentiment
Big business is getting bigger. Since the 1990s, America’s largest corporations have been growing larger and larger, and concentration is seemingly up in most markets.30 As a result, there are fewer but bigger large corporations today, and they encompass a larger fraction of the economy than ever before.31
After three decades of increasing size and concentration came the public backlash: a reawakening of a strong anti-bigness sentiment that is deeply rooted in American culture.32 As a result, antitrust law is now back in vogue, reaching the top of the public, media, and political agendas.33 President Biden signed an Executive Order that commits the entire federal government to reining in the power of big business.34 The President also appointed noted anti-monopolist critics to key positions: Lina Khan to head the Federal Trade Commission (FTC), Tim Wu to a post in the National Economic Council, and Jonathan Kanter to lead the antitrust division at the Justice Department.35 Lawmakers have introduced a series of bills “representing the most aggressive effort to remake antitrust law in a century.”36 And law enforcers have filed detailed lawsuits described as “the government’s most significant [legal] challenge” to market power in a generation.37
This strong anti-monopoly sentiment does not rely on harms to consumer welfare as its rallying cry. After all, unlike the monopolists of old, today’s superstar companies do not charge high prices or restrict output.38 Google and Facebook, for example, supposedly charge nothing, and Amazon and Wal-Mart compete by keeping prices permanently low.39 Indeed, many scholars argue that today’s super-large companies became so big precisely because they offered better, more innovative products; and that fewer larger firms can and do compete with each other more ferociously than many smaller firms do.40 The anti-monopoly camp has offered several counter-rebuttals.41 One set of rebuttals stresses that lower prices do not necessarily mean more consumer welfare.42 With big tech, for example, super-large sellers enjoy greater capacity to redistribute wealth from buyers to themselves without increasing prices, such as by exploiting consumer data and attention.43 Another set of rebuttals shifts focus from how the new monopolies treat their consumers to how they treat their competitors.44 That way, instead of examining just near-term effects on consumer pricing, we can decipher what the future market will look like.45
But from our perspective, the most interesting set of rebuttals is the one that goes beyond competitive effects. The new school of anti-monopolists (dubbed neo-Brandeisian antitrust by its proponents and hipster antitrust by its critics)46 has shifted focus to how bigness causes many societal harms that cannot be measured by product prices alone, such as harms to our civil liberties, privacy, or the free press.47 In other words, the new critics of bigness emphasize not only the economic consequences of bigness, but also how it impacts our democratic institutions writ large.48
Against this backdrop emerged the “big is ungovernable” claim. A clear manifestation of “big is ungovernable” came in the abovementioned 2020 congressional report on big tech.49 The claim has since been frequently invoked by anti-monopolists, reiterating that “for the most powerful corporations, laws are often mere suggestions.”50 There is no point in trying to order them to respect user privacy or refrain from censoring speech; the only way to get these giant corporations to follow rules is by breaking them up, the argument goes.51
Yet up to this point, the “big is ungovernable” claim has remained underdeveloped. Are claims such as the ones appearing in the congressional report based on statistical evidence or mere anecdotal observations made by disgruntled rivals? Even if we assume that these claims are representative of a recurring pattern of behavior, who is to say that this pattern applies beyond the dominant tech platforms and across sectors to big pharma, big ag, and big banks as well? And even if we assume that most big corporations repeatedly ignore laws and court orders, who is to say that they are not accountable to societal needs and demands via other, nonlegal mechanisms, such as wanting to maintain good reputation and legitimacy? Indeed, there is ample theoretical and empirical analysis suggesting that larger corporations are more governable than small and midsized ones. Could it be that the “big is ungovernable” notion is merely riding the anti-bigness sentiment, with little actual merit?52 To examine this question, we need an organizing framework.
B. How Size Affects Governability: A General Framework
1. The Conventional Wisdom: Bigger Targets with Deeper Pockets
In conventional economic analysis, big means better governability. The case for economies of scale in deterrence is twofold, following the basic components of the economic theory of deterrence:53 (1) higher certainty of detection, and (2) greater severity of sanctions.
Probability of detection supposedly increases with firm size because larger corporations are more closely watched by more sets of eyes.54 Larger corporations have more interactions with stakeholders: more employees, more buyers, and a bigger public footprint.55 With more interactions comes a higher likelihood that one of these stakeholders will notice a negative event.56 Additionally, the largest corporations tend to attract the most coverage by media outlets,57 corporate watchdogs,58 stock analysts, and sophisticated institutional investors.59 More information is available on larger corporations and more audiences are interested in learning about them.60 Bigger corporations thus have bigger targets on their backs.
Once misconduct is detected, larger corporations would also suffer larger sanctions, or so the argument goes. Their pockets are deeper, which makes them more likely to pay larger legal sanctions in both public and private enforcement proceedings.61 And they experience a larger reputational fallout following bad news, if only because they have more preexisting reputation to lose.62
Conventional wisdom thus suggests that big business is more amenable to deterrence than smaller market actors, which are more obscure and have shallower pockets. Yet this conventional wisdom ignores two strong countervailing factors, namely, power and fragmented knowledge.63
2. Power
Any analysis of misconduct by big business should account for the ability of powerful corporations to influence what counts as misconduct to begin with. The conventional argument assumes that misbehaving corporations are “price takers” in the sense that they treat the certainty and severity of sanctions as givens. Such an assumption may hold for individuals and small or mid-sized firms, which lack the power to shape the laws and societal norms that govern them. Super-large firms, by contrast, can get away with (and even get praise for) behaviors that would be condemned when engaged in by smaller market actors. They not only game the rules but also rule the game.
To illustrate, a stylized fact in the empirical literature on corporate corruption is that “strong firms use their influences to change laws and regulations, whereas weak firms pay bribes to mitigate the cost of government intervention.”64 Any empirical study that measures “corporate misconduct” as actions that were prosecuted and punished should therefore be taken with a grain of salt: its results could reflect not necessarily the likelihood that larger corporations engage in more/less misconduct, but rather the likelihood that they will be targeted by enforcers.65 In other words, such studies fail to capture the ability of big corporations to engage in socially harmful behavior without being treated as wrongdoers. Indeed, organizational scientists have long emphasized that the line between right and wrong corporate behavior is fluid and depends on the judgment of legal and market arbiters, such as regulators and opinion leaders.66 These “social control agents” often label actions differently as a function of who the actor is.67
Super-large corporations are better at influencing social control agents and at operating in gray areas and exploiting loopholes. They can shape how their conduct is labeled by using their “power” straightforwardly, as in pouring limitless resources into lobbying the regulators or running public relations campaigns and paying influencers. But in many instances, these corporations do not even need to exercise power proactively: size connotes success, and their aura of success often suffices to shape the labeling of their conduct.68 For example, one account of misconduct in the financial sector highlights how Goldman Sachs’s shenanigans often result in the firm getting credit for being the smartest and pushing the envelope, whereas similar behaviors by smaller actors are condemned.69 Generally, when the biggest, most powerful actor in the industry engages in a certain behavior, we tend to deem it as legitimate, assuming that everybody else is/should be doing it.70
When social control agents nevertheless identify what super-large corporations did as transgressing the line, these corporations can still leverage their power to dilute the expected sanction for misbehaving. Here as well, power could be exercised proactively, as in forcing their counterparties to sign gag clauses.71 And power could also manifest passively, as in reducing the chance that a worker or a supplier would blow the whistle, for fear of higher costs of retaliation.72
3. Fragmented Knowledge
Larger corporations “tend to be more complex and decentralized.”73 Tasks are divided into numerous subtasks, and control and responsibility are dispersed among numerous subunits.74 The larger the corporation, the more segregated information flows become, such that no single individual can truly know everything material that is going on within it.75
Fragmentation of knowledge creates plausible deniability.76 Field work in large corporations has shown that knowledge of unethical behavior tends to be pushed down the organizational ladder, while credit for success is claimed up the ladder.77 The larger the corporation, the less likely it is that damning information would flow vertically up the hierarchical chain, or horizontally across different units.78
Plausible ignorance hinders the effectiveness of all systems of control. Legal sanctions in the corporate arena usually hinge on proving awareness, and fragmented knowledge makes it harder to disprove ignorance.79 Moral sanctions stem from anticipatory guilt for misbehaving. Yet if someone is not aware of the moral consequences of their company’s actions, they are protected from self-condemnation and will be able to sleep better at night.80 Reputational sanctions hinge on attribution, namely, how the company’s stakeholders assign the blame of what happened.81 The more fragmented knowledge and responsibility are, the harder it becomes for outsiders to pin blame and evaluate whether past behavior is indicative of future behavior.
Plausible ignorance could be the product of deliberate action. That is, large corporations may deliberately divide tasks and erect barriers to information flows in order to create “distributed ignorance.”82 Yet calculated maneuvering is not necessary in super-large corporations, as plausible deniability emerges there organically.83 As corporations grow in size and complexity, they have to make tough decisions on how to manage the information flows between their numerous subunits. Seemingly benign features of managing the information flow, such as division of labor and specialization, then create information silos, which eventually lead to undesirable (from a broader societal perspective) consequences.84 Whether such maneuvering is deliberate or not, the upshot for our purposes is the same: size leads to fragmentation of knowledge, and fragmentation of knowledge reduces the ability to detect and stop misconduct. Indeed, post-mortem analyses of debacles in giant corporations usually reveal that several insiders flagged the problems in real time, but the damning information was too scattered across units for the organization to follow through on it.85
***
Thus far, we have discussed the vectors linking size to governability in abstract terms. In reality, none of these vectors are absolute or deterministic. Rather, the pull of each vector varies across contexts and institutions. For example, the plausible deniability dynamics are more relevant for law enforcement than for market discipline and, within law enforcement, more relevant in the criminal context than in private litigation. And the aura of success of super-large corporations may help them get the initial benefit of the doubt (when market actors assess their blameworthiness), but hurt them when it is clear that they violated the higher expectations society has of them.86 It is therefore time to inject more context and apply the general framework across different systems of control.
II. Laws
In current debates, the two most familiar variants of the “big is ungovernable” claim are “regulatory capture” and “too big to jail.” The former suggests that big corporations pour money into lobbying and campaign donations to capture the regulatory framework that is supposed to govern them. The latter maintains that prosecutors are reluctant to vigorously pursue the toughest sanctions against giant corporations for fear of systemic collateral damages. This Part evaluates these popular claims and fleshes out different, more obscure mechanisms through which bigness can dilute legal deterrence. Section II.A emphasizes big business’s ability to influence the regulatory framework not just via lobbying but also via subtler, under-the-radar methods. Section II.B explains how fragmented knowledge makes it harder to investigate and prosecute top individual decision makers, thereby diluting public enforcement; and how market power allows giant corporations to force their counterparties to waive their private rights of action, thereby diluting private enforcement.
A. Regulation
[T]he growth in the platforms’ market power has coincided with an increase in their influence over the policymaking process. Through a combination of direct lobbying and funding think tanks and academics, the dominant platforms have expanded their sphere of influence, further shaping how they are governed and regulated.87
For our purposes, “regulatory capture” refers to the ability of corporations to push regulators to alleviate certain burdens on themselves or impose new burdens so as to make it harder on their competitors.88 When examining how size and market power affect the ability to influence regulation, there exist two contrasting theoretical arguments.
On one hand, bigness mitigates the collective action problem that plagues political influence efforts.89 Mancur Olson famously observed that political influence is, in a sense, a public good: firms that invest in lobbying for their favored policies cannot exclude others in the industry that did not pay for lobbying from reaping its benefits.90 This creates incentives to free ride on political influence efforts. Size and market power are classic antidotes to the free-rider problem.91 For enormous companies, the individual benefits from obtaining beneficial regulation are large enough to justify investing in lobbying regardless of what others do.92 And in concentrated industries, the relevant group consists of a smaller number of members, thereby lowering the costs of coordinating for action. Industries with a few large firms will be more effective in lobbying than industries with a large number of smaller firms.93 The result may be a vicious circle: once a firm gains economic power, it can employ its vast resources to purchase political power, capturing the regulation to fend off competitors, which, in turn, further buttresses the firm’s market power, and so on.94
On the other hand, super-large corporations are more publicly visible, making catering to them riskier for policymakers. Elected officials or professional regulators have much to lose from being perceived by the public as catering to big business, and much to gain from regulating them vigorously, being perceived as protectors of the little consumer/employee/citizen.95 Besides, what matters in politics is not just mobilizing material resources (as in lobbying expenditures), but also legitimacy and having a compelling narrative to justify your preferred policies in the court of public opinion.96 And in this battle—the battle over legitimacy and narratives—size can be a disadvantage. The public is likely to be skeptical of and antagonistic to claims coming from big business.97 In other words, big business is too big to cater to.
Whether market power translates to political power is therefore an empirical question. Critics of the anti-monopoly movement often make two claims in that regard: (1) the relationship between firm size and political power is not borne out in the data;98 and (2) in any case, the data suggests that the overall amounts of money in politics are too low for us to consider them a game changer.99 Yet a more careful look at the evidence allows us to decipher when and how bigness leads to regulatory capture, along the following three dimensions.
First, the relevant empirical studies are those that do not look just under the streetlight. Clean large-scale data on political influence efforts is notoriously hard to collect.100 Researchers are thus often tempted to use proxies, such as the number of registered lobbyists or self-reported political donation expenditures.101 But while researchers gravitate toward relatively transparent means of political influence, savvy political actors (rent-seeking firms and rent-extracting regulators) gravitate toward the covert ones. The more transparent a political-influence conduit, the less representative it is of influence efforts as a whole.102 Therefore, to gain a sense of the link between market and political power, we need to locate empirical studies that bypass this methodological problem with creative designs.
One example comes from Russell Pittman, who took advantage of sudden changes in transparency: he examined certain types of political donations that were historically not subject to disclosure, until a later court decision made them retrospectively public.103 Tellingly, Pittman found a positive correlation between market power and investments in political donations: companies with more market power were pouring more money into this stealth political-influence channel.104 Recently, a super-group of economists spotlighted the role of corporate charitable contributions as a covert, tax-exempt form of political spending.105 For example, they found that nonprofits that receive donations from corporations tend to submit comments to the regulators in support of the position that favors the donating corporation.106 Pertinently here, the larger the corporation, the more likely it is to engage in such charitable contributions through its philanthropic foundations.107
Such studies illustrate the problem with the “there’s too little money in politics” argument. To be sure, if one looks solely under the streetlight (traditional PACs), it seems as if there is “little money.” But once one looks closer at darker corners (charitable contributions), the money stops being so little.108 These studies also explain why the “higher visibility” argument is overstated. The too-visible-to-cater-to argument loses traction once political influence is exercised through low-visibility channels.
A second way to bypass the methodological problem and assess the link between market power and political power is by looking at outputs rather than inputs. Instead of gauging how many resources companies invest in purchasing political influence, output-based studies examine whether larger corporations tend to get the regulation they want.109 One such study found that larger corporations enjoy more favorable tax rates.110 Another found that industries that are more concentrated enjoy more protectionism in the form of regulatory restrictions on imports.111 Perhaps most importantly, output-based studies find that big corporations tend to get what they want more at the industry-specific regulatory level than at the national level. Historically, most studies looked at general influence at the congressional level, focusing on legislative outcomes.112 Once researchers started examining more pinpointed influence on specific regulations, the evidence linking firm size to political influence became more pronounced.113 In other words, while big corporations may not control the broader political agenda, they have meaningful influence on the regulatory agenda.114
In fact, one could claim that even output-based studies understate the actual influence of big business. These studies focus only on observable outcomes, namely, decisions that regulators make on issues that are already on the agenda. Yet it is plausible that a key outcome of big business’s political influence is that of keeping certain thorny issues off the agenda to begin with.115 The existing studies would have a hard time measuring such regulatory omissions.
A third way to fully comprehend the link between market power and political power is therefore to focus on more implicit channels of influence. Regulation need not be directly “acquired by the industry,” as George Stigler famously quipped,116 in order for it to de facto serve the interests of big corporations at the expense of a dispersed public. Regulatory capture in developed countries with a strong rule of law rarely comes from explicit bribes or threats.117 It rather comes from subtler, lawful influence channels, such as revolving doors between the public and private sectors, “epistemic capture” due to information asymmetries and complexity, or “cultural capture” due to regulators’ identification with the companies’ worldviews.118
And the important point for our purposes is that stealthy political influence is likely to increase with firm size, for various reasons.119 Most basically, larger firms can offer greater potential payoffs to policymakers who play along with them, even without an explicit quid pro quo deal taking place. For elected officials, catering to super-large corporations (while cloaking it as promoting the public interest) could mean gaining a critical mass of votes.120 For professional regulators, it could mean lucrative future business opportunities.121 Indeed, when damning information about C8 emissions started getting out locally, DuPont hired the former local West Virginia regulator; and when the bad news made it to the national stage, DuPont hired two former Environmental Protection Agency (EPA) higher-ups to handle its communications and regulatory affairs.122 Generally, political engagement comes with declining marginal costs, and super-large corporations can more easily pay the high fixed costs of becoming truly influential: not just the costs of establishing a government affairs department and political action committees, but also the costs of building long-term relationships over time.123
Further, size comes with complexity, and complexity is an important determinant of regulatory capture.124 Complexity reduces the threat of public backlash against capture, as nonexpert citizens are less likely to keep score.125 Complexity also makes it more likely that regulators would over-rely on regulated entities in order to make sense of the overwhelming amount of information they have to process.126 To illustrate, when damning news about C8 broke, DuPont defended itself in the courtroom and in the court of public opinion by arguing that the relevant regulators knew about C8 but did not order the company to stop emitting it.127 In reality, what the company did was submit lengthy documents to the regulator where they buried some damning details about C8 (often called by different names) along with detailed reporting on hundreds of other chemicals.128 Larger corporations can more easily hide the ball that way: they bombard regulators with a barrage of unprocessed, hard-to-comprehend information, making it less likely that they will be regulated effectively.129
A firm’s size and market power also increase its ability to influence regulation via “cultural capture.”130 Cultural capture denotes instances where regulation systematically caters to special interests not because of quid pro quo, but rather because regulators truly (but overly) identify with the industry’s worldviews.131 Identification could come from these firms’ abovementioned aura of success. Or it could come from their ability to fund research that “wrap[s] its self-interest in a bigger, noble, idea.”132 Indeed, the congressional report on big tech provided multiple examples of the dominant platforms funding think tanks and nonprofit advocacy groups to shape the public’s (and policymakers’) opinion.133
In sum, one cannot dismiss the link between market power and political power. Big business has a decided advantage in influencing regulatory decisions on issues with low saliency and high complexity, where decisions are made away from the public eye. At the same time, regulatory capture is hardly absolute or deterministic.134 The largest corporations do not always have their way, and they do not infrequently find themselves in violation of legal requirements. The question then becomes, how effective is law enforcement in detecting and punishing such violations?
B. Litigation
“[C]ourts and enforcers have found the dominant platforms to engage in recidivism, repeatedly violating laws and court orders.”135
Many academic analyses assume that big corporations are actually more likely targets for law enforcers due to their higher visibility and deeper pockets.136 A corporation with $500,000 in assets that causes environmental harms with cleanup costs of $5 million cannot internalize the costs of the harm it causes. Super-large corporations, by contrast, rarely fall into such “deterrence traps.”137 Further, public enforcers, such as the Department of Justice (DOJ) or Securities Exchange Commission (SEC), cannot sweep allegations against big corporations under the rug, because doing so would danger the enforcers’ reputations.138 And private enforcers, such as class action lawyers, anticipate that targeting bigger defendants would come with higher fee awards.139
While the bigger-targets-with-deeper-pockets argument sounds intuitive, it cannot account for on-the-ground evidence showing that larger firms are more prone to recidivism.140 To illustrate, back in 2012, Facebook settled with the FTC regarding charges of deceiving consumers about monetizing their privacy and promised to amend its ways. Seven years later, while investigating the Cambridge Analytica scandal, the FTC concluded that “Facebook . . . almost immediately beg[a]n violating [the 2012] order following its adoption.”141 Similarly, an oft-ignored fact about the Volkswagen emissions scandal is that in the 1970s, Volkswagen had already been installing “defeat devices” to evade vehicle emissions regulations. The EPA caught and reprimanded Volkswagen back then, but that apparently did not deter the car giant from continuing with such misbehavior, and on a much greater scale.142
If super-large corporations know that their misconduct is more likely to be detected and heavily punished, why do they keep breaking the law and exposing themselves to potential punishment? Part of the answer probably stems from the costs of internal compliance: in complex corporations with dispersed responsibility, it is harder to ensure that everybody follows the law.143 Another part of the answer are the costs of external enforcement against big corporations, and the perception that these corporations can get away with profiting while ignoring the law. This is where we move to now.
1. Difficulties in Detecting Culpable Individuals
Not a single top executive has been prosecuted in the largest corporate debacles of the past decade, such as the opioid crisis or GM’s faulty ignition switch.144 The reason has to do with fragmented knowledge. Bigness reduces the chances of proving awareness.145 In small or mid-sized corporations, top managers tend to be involved in all key aspects of the business; in super-large corporations, by contrast, top managers tend to be many steps removed from actual wrongdoing.146 Indeed, legal scholars have long acknowledged that criminal law’s settled concepts, such as the mens rea (or state of mind) requirement, do not fit well with the modern giant corporation.147
Beyond aggravating the difficulty of proving knowledge and intent, bigness also dilutes the prospect of individual accountability through the ability of giant corporations to dispense limitless resources to help their top decision makers fight off allegations or enter agreements that settle these allegations.148 In the Cambridge Analytica scandal, for example, Facebook was accused of agreeing to pay the FTC a much heftier fine ($5 billion) in exchange for the regulator waiving personal liability for Mark Zuckerberg and other Facebook higher-ups.149
On paper, we can solve problems with prosecuting individuals by imposing liability at the entity level. That is, instead of policing individuals, enforcers can police the giant corporations themselves, which would in turn incentivize corporations to police the individuals within them.150 But as the Facebook example just illustrated and Section II.B.2 elaborates, entity-level sanctions are unlikely to faze super-large entities because real-world sanctions are often too capped and lagged.
2. Difficulties in Properly Calibrating the Sanction
It takes massive sanctions to deter massive corporations. After all, the size of the fine should fit the social harm done by the wrongdoer, and larger corporations typically have the capacity to inflict greater harms.151 To illustrate, in Volkswagen’s emissions scandal, eleven million cars were polluting, and in Wells Fargo’s phony-accounts scandal, three and a half million fraudulent accounts were opened.152 Misconduct on such a scale should be subject to fines that are proportionately larger: not just bigger per se, but that much bigger.153 If the bigger firm’s wrongdoing inflicts social harm ten times the magnitude of that of a smaller firm, the fine should be ten times higher.
Yet for various reasons, actual sanctions are capped.154 Sometimes enforcers are bound by the statutory framework itself. Criminal fines, in particular, come in statutorily fixed amounts.155 Other times, enforcers can, on paper, impose huge fines, but they do not want to for fear of collateral consequences.156 Criminally prosecuting corporations could lead to severe sanctions, such as delicensing large professional service providers, debarring large government contractors, or not allowing large pharmaceutical companies to sell drugs through Medicare and Medicaid.157 Public enforcers fear that such measures would cause even giant companies to tumble, which would penalize innocent stakeholders.158 Size monotonically increases the number of innocent bystanders that would suffer: more public shareholders, workers, suppliers, and so on. And above a certain bigness threshold, companies become systemically important, such that the collateral damages extend beyond their stakeholders and spill over to the entire economy; hence the “too big to fail” moniker.
A classic example comes from the collapse of accounting giant Arthur Andersen following its Enron debacle–related indictment. The Arthur Andersen example is apparently still fresh in public enforcers’ minds, limiting their willingness to proportionately calibrate sanctions.159 Indeed, when former United States Attorney General Eric Holder was asked why he had not indicted the largest financial institutions for their wrongdoing leading to the 2008 financial crisis, he admitted:
I am concerned that the size of some of these institutions becomes so large that it does become difficult for us to prosecute them when we are hit with indications that if we do prosecute—if we do bring a criminal charge—it will have a negative impact on the national economy, perhaps even the world economy.160
Beyond fear of collateral consequences, public enforcers are also reluctant to vigorously pursue cases against the biggest corporations because they seek to preserve their department’s scarce resources and their own reputations.161 The objective difficulty of disproving ignorance within super-large firms is combined with the fact that these defendants spend limitless resources on fighting their cases.162 Public enforcers thus anticipate lengthy battles in court that would drain their resources and come with a relatively low win rate.163 And because public enforcers are risk averse with respect to resources and reputation, they tend to avoid such high-risk–high-reward battles.164 Instead, enforcers prefer quick small wins: prioritizing going after misconduct by small rather than big market actors, and attempting to prove strict liability–type offenses rather than negligence or awareness.165
When they do pursue cases against giant corporations, public enforcers prefer pre-negotiating settlements in which these corporations pay fines that are indeed larger than the ones paid by smaller corporations, but not proportionately calibrated.166 Such settlements are good for both parties. Enforcers can celebrate collecting large fines and direct their limited resources elsewhere, in a continuous effort to ramp up the observable yardsticks of the number of cases brought and amount of money collected.167 The misbehaving companies get to write off the fine as the cost of doing business. Tellingly, announcements of non-prosecution or deferred-prosecution agreements are often met with positive stock market reactions, even when the company agrees to pay a ten-digit fine.168 The market apparently realizes that post-settlement monitoring is rare, as enforcers prioritize working new cases over policing old agreements.169 Indeed, when Facebook announced the abovementioned $5 billion fine with the FTC, its stock price jumped up significantly.170 And in the C8 case, the EPA touted a $16.5 million settlement with DuPont as “the largest fine ever collected” for such misconduct, while a rough cost-benefit analysis shows the huge difference between “largest ever” and “large enough to deter.”171 In order to deter it from continuing to emit C8, DuPont should have expected a sanction north of $1 billion.172
And even when huge sanctions are imposed on giant corporations, their deterrent effect can be severely diluted because of the time lag. There is often a large delay between when corporations engage in misbehavior and when they have to pay for it. With DuPont, the key decision to continue polluting without abating came in 1984, while the significant sanctions only started coming in 2017: a thirty-three-year time lag. Economist Luigi Zingales and I demonstrated that once you calculate the time value of money, the decision to pollute becomes rational from the point of view of shareholder wealth maximization.173 Roughly speaking, a $1 billion sanction in 2017 is the equivalent of $100 million in 1984.174 It paid to pollute because of the huge time lag between polluting and paying for it. The bigger the corporation, the bigger the time lag: enforcers need more time to properly investigate and locate culpability, and defendants have more resources to invest in delay tactics.
In sum, while decision makers in small or mid-sized corporations anticipate that blatant, socially harmful misconduct could cost them their business or their liberty, decision makers in super-large corporations can discount that risk, knowing that, at worst, their company will pay some diluted fine down the road. In a way, this is the mirror image of the deep-pockets argument. Larger corporations can more easily absorb legal expenses and punitive awards. They therefore treat legal sanctions as part of the larger costs of doing business. Deeper pockets translate to effective deterrence only when sanctions are properly calibrated. When sanctions are capped and lagged, deeper pockets translate to ineffective deterrence.
3. Contracting Out of the Legal System
We have focused thus far on the limits of public enforcement. Yet private enforcement could be, on paper, better geared at deterring misconduct by massive corporations. Certain features of U.S.-style private litigation—extensive discovery, outsized punitive damages, and high attorney fees in class actions—make larger defendants better targets for private attorneys looking to maximize their earnings. Plaintiff attorneys can credibly threaten to impose the direct and indirect costs (time, embarrassment, reputational costs) of discovery, as well as the threat of larger punitive damages (which increase in defendant size), to extract quicker and heftier settlements when targeting bigger corporations in class actions.175
In reality, however, giant corporations have managed to dilute the effectiveness of private actions as well, by forcing their counterparties to sign mandatory arbitration provisions, class action waivers, and gag clauses.176 In a string of rulings over the past decade, the Supreme Court gave its stamp of approval to such provisions.177 Companies did not waste time installing them in their contracts with consumers, workers, and suppliers, so that nowadays, if you purchase a phone or start a new job with a large corporation, chances are you are subject to one.178
Pertinently, the ability to force these provisions is a function of the firm’s size and market power. To illustrate, consider the chicken industry example that is the focus of Zephyr Teachout’s 2020 book.179 Teachout notes that over time, a small number of distributors (think Tyson or Perdue) have taken control of almost all aspects of chicken production.180 She then ties the consolidation in the market to the proliferation of mandatory arbitration and gag clause provisions: chicken farmers have no choice but to sign these waivers, and so they now cannot air their grievances in a public courtroom and cannot even tell others that they were abused.181 Had the chicken industry not been controlled by a few super-large distributors, such provisions would have been far less likely. This is therefore a live example of how market power translates to diluting the threat of legal sanctions.
Another doctrinal development from the previous decade that reduced the ability to employ class actions against super-large companies came in Wal-Mart v. Dukes.182 When one and a half million female Wal-Mart employees asked the court to certify their employment discrimination class action, the Supreme Court refused, overturning the Ninth Circuit and heightening the “commonality” requirement of Rule 23 of the Federal Rules of Civil Procedure.183 The Dukes court reasoned that, in a giant corporation such as Wal-Mart, decision-making is highly decentralized, so the employees’ discrimination claims would not have enough in common to join in a single suit.184 Labor law experts noted in real time that the decision “effectively reinforced the advantages of corporate scale,” helping super-large corporations immunize themselves against legal action through “the very same feature that makes [them] so powerful to begin with.”185
With class actions losing relevance, the onus is on individual litigation. Yet here, the ability to discipline super-large corporations is limited by inherent power imbalances. Our adversarial litigation system is based on the notion of a level-playing-field competition that produces just and truthful outcomes.186 But the level-playing-field metaphor is completely inapt in light of the massive gaps in resources, abilities, and incentives that exist when individuals who are harmed sue the super-large corporations that harmed them.187 Unsurprisingly, evidence suggests that most individuals forego pursuing their rights against super-large corporations to begin with.188
In sum, there is ample reason to believe that super-large corporations can discount the expected legal sanctions for misbehaving. Still, they could have other, nonlegal incentives to invest in the quality of their products, the safety of their workplaces, and abatement measures.189 The next two Parts examine how size affects these nonlegal sanctions and rewards.
III. Reputation
Big corporations care deeply about building and maintaining a strong reputation among their consumers, workers, suppliers, nearby communities, and regulators. Accordingly, even if one assumes that super-large corporations capture the regulatory framework and are too big to investigate and punish, it does not follow that they are ungovernable. Indeed, classic reputation models suggest that larger corporations are more likely to behave according to market and societal norms for fear of reputational fallout.
This Part evaluates this conventional wisdom and finds it wanting.190 The classic models emphasize how larger firms have more to lose from reputational sanctioning, but underplay the other side of the equation, namely, the higher costs of “punishing” larger firms. Punishment here means loss of future business opportunities: your stakeholders switching to your competitors. Yet when the firm in question is super-large with market power, severing ties with it becomes harder. Section III.A highlights studies corroborating this point. Section III.B examines how big corporations with market power can block damning information about them from getting out. Section III.C explores how these corporations can tarnish the credibility and manage the meaning of whatever damning information ends up getting out.
A. The Conventional Wisdom and Its Flaws
Unlike in the legal literature, where there exist prevalent arguments against bigness, within the reputation literature, the prevalent argument is that big is better. Large, established firms have a big reputational cache to protect and would therefore invest their vast resources in internal compliance programs. When a firm sells many products, its entire reputation is at stake whenever a single product is sold. Therefore, the more products a firm sells, the more valuable its reputation for product safety becomes.191 In economic lingo, this means that larger firms post larger “reputational bonds.”192 Beyond larger corporations having more to lose, classic reputation models also suggest that these corporations face higher certainty of detection because probability of detection is positively correlated with the frequency of repeat play, and larger firms play more repeatedly.193
This reputation economies of scale argument, however intuitive, fails to account for two important dynamics. First and most basically, the argument focuses on how much large firms have to lose from violating market norms, but it neglects how little stakeholders have to gain from “punishing” large firms with market power. Reputational sanctioning is the process of stakeholders hearing bad news about the firm, downgrading their beliefs about it, and switching to competitors. Concentration reduces the number of viable alternatives, thereby increasing the costs for market actors of switching to competitors.194
A recent empirical study illustrates this point by examining how news about misbehavior by retail outlets affected consumer willingness to patronize the outlets going forward. The study found that the reputational sanction depends strongly on the level of competition in the market.195 Outlets facing less competition experience smaller drops in consumer visits following bad news. The authors conclude that “competition is an important component in customers’ ability to discipline firms for misbehavior.”196
A second fundamental flaw with the reputation economies of scale argument is that it treats the process of reputational settling up as a binary, one-sided affair, whereby once information about an adverse action by the firm becomes public, that firm suffers meaningful reputational damage. In reality, reputational sanctioning is a very noisy process.197 Similar behaviors often lead to different reputational outcomes: one firm emerges from a product recall unscathed, while another goes bankrupt. Damning information about the firm does not automatically translate to reputational damage.198 For meaningful reputational sanctions to occur, several other conditions have to hold. The information has to be widely diffused, perceived as credible, and attributed to deep-seated problems in order for the reputational sanction to be meaningful.199 Importantly, reputational sanctioning is hardly a one-sided affair. Super-large corporations pour resources into affecting each of the abovementioned conditions. Even if they fail to block information from getting out, they can limit the diffusion, tarnish the credibility, or manage the framing so as to limit attribution to the corporation.200
B. Blocking Information from Getting Out
The prospect of reputational discipline hinges on damning information about corporate misbehavior coming out. Information on corporate fraud comes from a plethora of sources, such as whistleblowers from within the firm, and journalists and short sellers from outside the firm.201 Size and market power can make it harder on these sources to uncover and disseminate damning information.
Consider whistleblowers first. Size increases the fragmentation of knowledge and thus decreases the likelihood that any given employee would be fully aware of the misconduct. So, while there are more potential whistleblowers in bigger corporations, each of them would have a much harder time knowing if, when, and on what to blow the whistle. Market power, in turn, increases the costs of blowing the whistle. The more concentrated an industry is, the bigger the threat of retaliation becomes. Potential whistleblowers in concentrated industries anticipate that blowing the whistle would lead to burning bridges with their current company, leaving them with few alternative employers to switch to in the industry they specialize in.
A vivid illustration comes from the cheerleading industry, where a company called Varsity Brands had apparently monopolized the sport.202 In 2020, USA Today exposed systemic sexual abuse in the cheerleading world, whereby dozens of convicted sex offenders were allowed to continue working.203 Matt Stoller suggested that this total breakdown of self-regulation was the product of bigness: industry insiders told him that they were afraid to speak out because once they lose their current job, they would have no alternative firm to work for.204 The claim is that monopolization made the industry ripe for such scandals. In the DuPont case, several insiders similarly admitted after the fact that they kept quiet because, as one worker put it, “I wasn’t about to go against the paycheck that supported my family. So I shut my mouth.”205
If company insiders either do not know enough or know enough but are too afraid to blow the whistle, perhaps the victims of corporate misconduct would. We would expect consumers who purchased a faulty product, or nearby community members who drank polluted water, to air their grievances in public, putting a reputational sanction in motion. Yet here as well, there are significant costs for telling the world about misconduct by powerful corporations. The congressional report on big tech is filled with testimonies of market actors who were afraid of speaking out against abusive and discriminatory behavior by the big platforms for fear of retaliation.206 Peer pressure could also be at play, especially when the super-large corporation is the biggest employer and benefactor in town. The community members who approached journalists and took DuPont to court over C8 emissions were treated like lepers in their communities. They had bottles thrown at them and were forced to switch churches several times.207 Similar dynamics appeared in other cases, such as in the Grace Corporation’s asbestos pollution scandal in Montana.208
What about journalists and NGO watchdogs, then? The conventional wisdom is that information intermediaries are more likely to focus on the largest corporations.209 But there exists a key distinction between misconduct that has already been flushed out and misconduct that is still under wraps. Empirical studies pointing to greater media scrutiny of larger corporations focus on instances where companies already admitted to wrongdoing, such as by restating their financial reports.210 These studies do not focus on the question that is more pertinent here, namely, whether journalists are likely to reveal damning information that the company is trying to conceal. In other words, existing studies reveal a piling-on effect: once bad news is out, it makes sense for journalists to focus on the more newsworthy transgressions of larger corporations.211 The studies do not tell us, however, that journalists are more likely to extract damning information from super-large corporations to begin with. In fact, communication scientists point to several factors that hinder the effectiveness of media scrutiny of big business.
To extract damning information that the corporation attempts to keep under wraps, journalists need sources. One valuable source for investigative reporting is insiders—but here, our previous discussion of the difficulties of blowing the whistle against large corporations applies. Another valuable source for such investigative reporting is the legal system. In a separate project, I analyzed the content of prize-winning investigative reports over the past twenty years, showing that the majority of them rely heavily on “legal sources,” such as documents exposed in discovery or regulatory investigation reports.212 Yet the discussion in Section II.B.3 showed that big corporations can disable such legal sources by, for example, forcing their counterparties to sign mandatory arbitration and gag provisions. The ability to keep legal disputes away from the public eye dilutes the prospect of reputational deterrence, as it keeps damning information away from journalists.213
Even when journalists can gain access to damning information about the largest companies, they may not want to follow through on it. One well-documented channel for media slant is dependence on advertising revenues.214 A profit-minded media outlet would think twice before producing biting watchdog-type reporting on its biggest advertisers.215 Size and market power make this potential media slant worse, as bigger corporations in more concentrated industries are significant sources of advertising revenues. To illustrate with our recurring example: when victims of DuPont’s C8 emissions first learned about the dangers, they tried to enlist the help of local journalists, only to find out that local media refused to run the story against the biggest advertiser and employer in the community.216
C. Managing the Meaning of Information That Does Get Out
Beyond blocking damning information about them from getting out, super-large corporations can manage the meaning of information that does get out, in a bid to limit the chance that stakeholders will believe it and act on it. In other words, big corporations can dilute the reputational sanction by casting doubt on the veracity of the information or framing past events as not indicative of the company’s future behavior.
A burgeoning literature details the ability of large corporations to “bend science”217 and “manufacture doubt.”218 In the DuPont case, one of the documents that surfaced post mortem was the “Weinberg Memo,” in which a consulting firm painstakingly detailed a strategy to tarnish the credibility of scientific evidence on C8’s dangers.219 Further, several researchers testified that they were pressured not to publish damning results about C8’s effects in real time.220 Such an ability to dilute the certification of damning information requires heavy investments that only big corporations can afford: from employing a specialized PR division, to having all the top experts in the field on retainer, to funding “think tanks and nonprofit advocacy groups to steer policy discussion.”221
Even after damning information about them has been revealed, diffused, and certified, large corporations can try to limit the attribution of blame. Here, one might think that bigger corporations are at a disadvantage. Due to well-documented biases, stakeholders could perceive every bad thing that happens in the corporation as intentional and symptomatic of a toxic culture.222 And the media would be happy to highlight the negatives in order to cater to the anti-bigness sentiment.223 Still, size can also help the blame-shifting efforts in various ways.224 Consider the following two examples.
First, larger corporations can more easily utilize their preexisting corporate social responsibility (CSR) image as reputational insurance. When stakeholders perceive a company as generally socially responsible, they are more willing to give it the initial benefit of the doubt when allegations of misconduct surface.225 Empirical studies find that size increases this reputational insurance function of CSR.226 From a societal perspective, this could represent a net loss, to the extent that larger corporations’ engagement with CSR dilutes the expected reputational sanction and makes them feel freer to behave irresponsibly in other domains.227 A classic illustration comes from Enron, which, right up to its collapse, was widely considered a poster child for the CSR movement, winning many accolades.228 For some commentators, Enron’s strong CSR image contributed to the delay in revealing and punishing the company for its fraudulent activities.229
Second, beyond investing in CSR, the largest corporations often enjoy the benefit of the doubt simply because of their aura of success. As Section I.B.2 noted, market actors are more likely to interpret shady activities of super-large companies not as evidence of their disregard for market norms, but rather as evidence of their superior capabilities.230
In sum, while conventional wisdom is right to point out that big corporations have strong incentives to protect their reputation, it underplays how they can protect it, not necessarily by behaving honestly, but rather by managing appearances.
***
The DuPont-C8 case perfectly illustrates the limits of reputational discipline. Internal documents revealed that DuPont’s decision makers were anticipating in real time a very large reputational fallout if details about C8 emissions were to end up leaking out.231 They nevertheless decided to continue emitting, perhaps counting on their ability to delay and dilute the fallout. In retrospect, they were right: reputational deterrence broke down both at the individual and at the corporate level.232 A large time lag drives a wedge between individual- and corporate-level reputation concerns. The crucial decision to pollute was made in 1984, yet by the time bad news broke a couple of decades later, most of the 1984 board members were either dead or retired, so their labor-market reputation concerns were moot.233 The few 1984 directors who were still working in the 2000s were not once mentioned in media coverage of the C8 debacle; at most, the media reached out to current executives for a response.234
At the company level, reputational sanctions are limited to begin with in this context of environmental pollution.235 To generalize, the reputational sanction for harming contractual claimants (such as inflating the financial reports) is much bigger than the reputational sanction for harming unspecified third parties.236 More specifically, the DuPont case illustrates how large corporations can dilute reputational fallout by spinning off blame.237 Before the case went to trial and attracted intense negative publicity, DuPont spun off the unit that dealt with C8 to a new company named Chemours.238 The spinoff was unlikely to spare DuPont from its legal liabilities,239 but it may have minimized its reputational liability. Indeed, DuPont constantly refers media inquiries on C8 to Chemours, refuses to comment further on the issue, and points out that it is no longer DuPont’s business.240 Such spin-and-duck strategy is not unique to DuPont. Union Carbide followed the same path after the Bhopal disaster: in 1994, the Indian unit was sold to a local partner, and in 2001, Union Carbide itself merged with Dow Chemical.241
IV. Morals
The idea of morals as an internal system of control that deters misbehavior is intuitive: if one anticipates feeling strong shame and guilt for transgressing, one would avoid transgressing to begin with.242 That is, people may avoid behaving in ways that conflict with their set of internalized moral standards, regardless of how much this behavior benefits them materially.243 The interesting question for our purposes is the extent to which such self-regulatory mechanisms operate effectively on individuals working in super-large business corporations.
Critics of the anti-bigness movement maintain that the propensity of big corporations to misbehave is merely an extension of the propensity of individuals to misbehave.244 Sure, they argue, big corporations occasionally engage in immoral activities, but this is only because they are made of humans, and humans occasionally lie, cheat, and steal.245 If anything, big business is less likely to violate moral norms because behaving unethically is bad for business.246 Big corporations are more likely to instill programs and procedures that actually meliorate their employees’ underlying moral flaws, or so the argument goes.247
Yet, researchers have long shown that the organizational context matters: it changes how moral constraints operate and promotes both good and bad behavior by individuals inside the organization.248 Here, we would add that within the organizational context, size matters: bigness increases the risk of moral disengagement, whereby individuals can behave badly without feeling guilty about it. This Part taps into the social psychology literature to show how size affects the two major sources of variation in moral deterrence, namely, individual differences and situational differences.249 Section IV.A details the individual differences: internal promotion tournaments inside large corporations may tend to select individuals with moral plasticity for top positions. Section IV.B examines the situational differences: the diffusion of knowledge and responsibility within large corporations may sever the cognitive links between behaving badly and feeling bad about it.250
A. Hiring and Promoting Individuals with Moral Plasticity
Some individuals are less amenable to self-regulation, in the sense that they can engage in immoral activities and harm others without feeling much guilt. Personality types such as psychopaths, Machiavellians, and narcissists are relatively less bothered by conventional morality than the rest of us.251 The question here is whether the hiring and promotion processes in big corporations weed out individuals with such personality traits. We have ample reason to believe that they do not. In fact, individuals who can buffer their moral anxiety are probably more adept at climbing the ladder in super-large corporations.252
To understand why, consider the nature of internal “promotion tournaments” in the hypercompetitive corporate world.253 To be selected for promotion out of many other able candidates in your tier, you have to meet the yardsticks imposed by top management and outperform your peers. These “selection pressures” tend to favor certain adaptive traits and biases, such as willingness to take risks, good self-promotion skills, and moral plasticity. The one candidate who outperformed twenty of his peers is likely to be the one who took boom-or-bust decisions and boomed, drew credit to himself while shifting blame to others, and did not bother himself much with how he achieved his short-term yardsticks. These adaptive biases tend to come together as a package: a cluster of traits such as being high on narcissism and testosterone, and prizing toughness over fairness.254 Pertinently, individuals carrying this package tend to adopt “a more utilitarian kind of ethical reasoning,” whereby the means justify the ends.255 Those who reach the top at large corporations are therefore relatively more likely to be “adept at stepping ahead of peers when the finish line nears without the baggage of doubt or guilt.”256 Even if the overwhelming majority of them fall short of psychopathy, quite a few have moral plasticity.
Bigness could potentially aggravate the tendency to promote individuals with moral plasticity because of the type of working environment it creates. Personalities such as narcissists, Machiavellians, and functional psychopaths tend to be good at disguising their moral deficiencies over short periods of times, but their mask wears off in longer-term relationships.257 Super-large corporations tend to have more atomistic working environments and more rapid turnaround, making it harder on colleagues to develop deep familiarity and build up trust with each other. Such an environment of constant change makes it harder to detect toxic personalities.258 Relatedly, promotion decisions in larger firms rely more heavily on meeting observable yardsticks and claiming credit, and less heavily on intimate knowledge of candidates. As a result, there is a larger premium on self-promotion skills and “organizational politics,” which matches nicely with the strengths of grandiose narcissists, Machiavellians, and functional psychopaths.259
To clarify, in many aspects, having decision makers who are more likely to ignore rules to “get things done,” take risks, and remain cool-headed can actually be a boon for the corporation itself.260 Yet our perspective here is a broader, societal one: having decision makers with ethical plasticity renders moral deterrence ineffective.
Still, we should be careful not to overstate our claim. The state of knowledge of managerial personalities and how they affect firm behavior remains incomplete.261 We know enough to dispel the notion that internal competition and “the tournament process” weed out immorality.262 Yet we do not have enough hard evidence to justify jumping to concrete policy implications.263 Let us therefore move to a second, better-researched conduit linking bigness to morality, namely, how the structure of large corporations makes it more likely that even good people will do bad things.264
B. Good People Doing Bad Things
Even individuals with strong morals will feel less bothered by anticipatory guilt under certain conditions. Bigness activates some of these conditions, thereby facilitating moral disengagement. To understand why, let us first introduce concepts such as “moral wiggle room,” “role morality,” and “slippery slopes,” which explain how misconduct starts, spreads, and persists within organizations.
Individuals often make decisions without a lengthy deliberation of the moral consequences, based on an intuitive judgment of what is right or wrong.265 Sometimes we shift to an intuitive thinking mode simply to ease the cognitive burden.266 Other times we opt to avoid information in order to leave ourselves “moral wiggle room”: we know that acquiring information may make us feel bad about our choices, so we refrain from acquiring it.267
Another well-documented phenomenon is “role morality,” whereby an individual subordinates her general moral responsibilities to what she perceives to be her role’s responsibilities.268 When working for an organization, the individual may be willing to do things that she would never do in her personal life. As one top executive put it, “What is right in the corporation is not what is right in a man’s home or in his church. What is right in the corporation is what the guy above you wants from you. That’s what morality is in the corporation.”269 Ford engineer-turned-academic Dennis Gioia has vividly illustrated this point in a post-mortem analysis of the Pinto debacle.270
“Slippery slopes” here conveys the idea that many large corporate debacles start with small-scale, grey-area infractions that gradually descend into clearly wrong behaviors.271 There is not one clear point in time when actors have to make a discrete choice between behaving morally or immorally. Rather, misconduct progresses in small steps, allowing all of those involved to remain unaware of the moral wrongness of what they are doing.272 When participants finally realize how bad things are, it is often too late for them to reverse course: their preexisting commitment and loyalty to others in the group make them conceal the prior misconduct from the outside world.273 In this later stage of coverups, actors rationalize to themselves that the damage is already done and irreversible (say, the pollutants are already in the air).274 The right thing to do at this point, corporate actors tell themselves, is help the group you are a member of get through this episode with as little damage as possible.275
All these dynamics seem to be more pronounced in larger corporations. Size and complexity increase the fragmentation of knowledge and with it the displacement and diffusion of moral responsibility.276 The larger the corporation, the more tasks are divided into discrete subtasks, which in turn allows those inside the corporation to more easily avoid the broader implications of their actions (more moral wiggle room).277 Size also makes the slope slipperier. The process of gross misconduct through gradual descent hinges on the degree of awareness: the slower an awareness of wrongdoing comes, the more the corporation slips down the slope uninterrupted.278 Higher-ups will become aware of bad behavior very late in the process, when the corporation is already deeply committed to the fraudulent behavior.279 Further, size and complexity come with a style of management that heavily relies on delegation to self-operating units, whereby the higher-ups set observable targets for each subunit and step back.280 This “management by numbers” method, which relies on implicit directives (“we’re not telling you how to beat the numbers, but you better beat them”), breeds unethical behavior.281
In sum, bigness creates “a recipe for organizational wrongdoing that will never trouble the conscience of anyone within the organization.”282 Higher-ups do not know about the wrongful actions of their subordinates. Subordinates believe that they have no choice and that they are doing what is right within their role.283 Both are slow to fully comprehend the moral consequences that their actions add up to. Whether by facilitating moral blind spots or by facilitating conscious disregard, bigness may therefore bring down another bulwark against corporate misbehavior precisely when society needs it the most.
***
V. Implications
What, if anything, can be done to narrow the gaps in the ability of laws, markets, and morals to control super-large corporations? Potential solutions fall into two broad categories. The first is to limit the size to which corporations can grow, such as through merger approval doctrines. These sweeping antitrust-type solutions come with clear advantages and disadvantages: they are simpler to execute but may overly reduce economies of scale. The second category of solutions recognizes that some level of bigness will always exists in market economies and focuses instead on pinpointed attempts to fix flaws in existing systems, so as to make them better adept to deal with bigness. This Part evaluates the desirability of oft-proposed policy measures within that second category and outlines our own proposals for deterring misconduct by big corporations.
Section V.A provides background by taking stock of the evidence to highlight areas in which bigness-control problems seem more or less pronounced. Section V.B highlights in particular the need to focus on the interactions between the different systems of control. Existing policy proposals tend to focus on one issue at a time (such as “impose tougher campaign finance and lobbying restrictions!”). The problem with such a piecemeal approach is that specific proposals that look good on paper may end up being malleable to big business’s ability to bend and work around the rules and dilute enforcement. Policymakers should strive to make the different systems interact with each other positively, such that one system’s advantages check and balance the other systems’ flaws. Section V.C evaluates the desirability of common policy proposals such as increasing the severity of sanctions. The main upshot is that proposals are less likely to succeed unless they directly address the two key problems with controlling bigness, namely, power and information. Section V.D proposes concrete ways to address the core problems. For example, regulators could address the power problem by making class action waivers and gag clauses unenforceable, and courts could address the information problem by recalibrating director oversight duties and shareholders’ rights to inspect internal company documents. Section V.E offers big-picture lessons on the two hot topics of antitrust and corporate purpose.
A. Taking Stock of the Evidence: Variation
Before moving to concrete policy implications, let us take stock of what we can or cannot say about the links between bigness and governability. The analysis thus far has made it clear that, unsurprisingly, we cannot unequivocally say that “big is ungovernable” or that it is more governable. We should therefore exercise humility and refrain from sweeping statements.284 It is better to focus on the variation: identify the areas in which big corporations systematically perform better or worse than small and mid-sized ones, and then work backward to understand what this variation reveals.
Big is seemingly better in certain aspects of corporate governance.285 Larger corporations have more diverse and independent boards of directors and provide better tools for public shareholders to hold company insiders accountable, such as annual board elections, majority voting, and proxy access arrangements.286 Big also seems to fare better in preventing occupational injuries and racial discrimination.287 On the other hand, big corporations seemingly fare worse on other issues, such as certain aspects of pollution.288 What explains this variation? At the risk of generalizing in an area that does not lend itself to that, let us offer a few unifying principles.
Saliency seems to be a key determinant. Super-large corporations seem to do relatively better on issues that are clear and salient. Once the line between good and bad behavior is clear, and it is relatively easy to hold transgressors accountable, the biggest corporations do a better job of conforming to the rules. Relatedly, the identity of harmed parties also seems to play a role. Bigness-control problems are seemingly attenuated when it comes to misconduct against shareholders, such as deceptive accounting or option backdating,289 and misconduct toward employees, such as safety violations and racial discrimination. By contrast, bigness-control problems are seemingly pronounced in misconduct toward third parties, such as polluting the environment.290
To illustrate, consider how even within one company, namely, DuPont, there existed variation. During the same period that it was emitting C8, DuPont led several environmentally friendly initiatives, such as voluntarily halting production of chemicals suspected of destroying the ozone layer and ending its work on nuclear weapons.291 And even within its handling of C8, there existed a stark difference between how DuPont treated results about birth defects in employees (swiftly, by taking all female employees off the unit), and how it treated results about C8 leaking to nearby communities (negligently, by continuing emissions and keeping information away from community members).
The issue-saliency criterion explains the difference between handling C8 and handling the ozone layer or nuclear weapons issues. C8 was a lab-made chemical, whose existence (not to mention ramifications) was a closely guarded secret among a relatively small group of industry experts. Further, the risks from opaque chemicals remain relatively invisible for many years, and even when they materialize, it is hard to assign accountability to specific individuals.292 The harmed-parties criterion explains the difference between handling employees and handling nearby communities. The former were contractual claimants, and the (deformed) faces of babies born to female employees were something that DuPont decision makers could not disengage from. The latter were faceless victims, and whatever harm occurs to them is probabilistic and latent (say, an increase in the probability that some would get cancer twenty years down the road).
Another way to think about the variation is that big business is deterred mostly by reputation and less by laws and morals.293 Big corporations are carriers of reputation. When transgressions are salient and easy to understand, big corporations anticipate a large reputational fallout, which incentivizes them to ensure compliance to begin with. By contrast, when the issue flies under the radar and harms are remote, big corporations anticipate a limited-if-any reputational fallout. Unlike remote third parties, shareholders and employees know when they were harmed and who harmed them, and they can take their business elsewhere. They present a greater reputational risk.
It is exactly in these circumstances, where market discipline fails due to market power, externalities, and information asymmetries, that we need laws and morals the most. Yet, as Parts II and IV explained, bigness makes establishing legal liability harder and moral disengagement easier. When reputational deterrence breaks, there is little to stop big business from pursuing the strategy that maximizes the bottom line, even if it is inimical to broader societal interests.
***
Another source of variation in the body of evidence linking size to misconduct is the measure of size being used. Many studies refer to size in a purely relative way: a finding that “larger” corporations are less likely to engage in misbehavior could simply mean that corporations with 200 employees are less likely to transgress than are those with 100 employees. I focus here, by contrast, on bigness: super-large corporations with market power. From this vantage point, there is not much of a difference between corporations with 100 employees and those with 1,000 employees. Indeed, several disparate studies show that “size” is not linked to misconduct in a strictly monotonic way: the very largest corporations are disproportionately more likely to engage in illegalities or political influence efforts.294 In other words, there is not much difference between small and moderate-sized corporations, but there is a huge difference between the moderately-large and super-large corporations.
To be sure, some governance problems do tend to increase monotonically with firm size, while others do so only past a certain threshold. For example, the difficulty of detecting culpability due to fragmented knowledge increases relatively monotonically: the more employees and subunits there are in the organization, the harder it becomes to detect knowledge and intent.295 By contrast, corporations’ ability to influence the regulatory framework or “bend the science” tends to become apparent only once they cross a certain threshold. In other words, a corporation with 1,000 employees is not ten times more likely to capture the regulatory process than one with 100 employees; but a corporation with 100,000 employees is much more likely to set up dedicated PR and regulatory affairs departments, fund think tanks, and have well-reputed experts on retainer.296 To generalize, out of the two general vectors, fragmentation of knowledge seems to increase with firm size relatively linearly, while corporations’ power to shape the environment they operate in tends to come only past a threshold.
A related distinction is that between concentration and size. Company X could be smaller than Company Y but have more market power, since the two operate in different markets. Some of the control problems, such as the costs for stakeholders of punishing the company by taking their business elsewhere, would be more pronounced with Company X, even though it is smaller (think of the cheerleading example).297 Other control problems, such as in the difficulty of proving awareness by top-level decision makers, would be more pronounced with Company Y, even though it has less market power. Social planners looking to address bigness-control problems should be mindful of these linear-versus-threshold and size-versus-concentration distinctions.
A good illustration of the nuance and variation in our ability to control bigness comes from a 2018 study of employment discrimination lawsuits.298 The study shows that larger, higher-status corporations were less likely to be found liable by juries when charged with discrimination.299 The authors interpret this result as a “halo effect” working in favor of larger corporations when juries need to assess their blameworthiness.300 Yet in the subset of cases where juries did find them liable, those corporations received harsher punishments.301 The authors call this a “halo tax” and hypothesize that it stems from the same forces that create the halo effect, namely, higher expectations.302 Applied here, that study illustrates the point about how social control agents define “misconduct” differently as a function of the actor, such that super-large corporations are less likely to be labeled as wrongdoers. At the same time, if the misconduct in question is clearly established and the harmed parties are clearly identified, larger corporations can expect bigger sanctions and would have stronger incentives to curb misconduct ex ante.
B. Highlighting the Interactions Between Different Systems
We saw that each of the systems—laws, markets, and morals—has its limitations when dealing with super-large corporations. Yet even if each system is deeply flawed, the combination of all three could meaningfully check the behavior of massive corporations, as long as each system’s flaws are imperfectly correlated with the others’ flaws. The most problematic circumstances are those where all systems of control break together (for example, extreme information asymmetries, such as in the C8 case, render all systems ineffective), or those where companies can use a flaw in one system to hinder the effectiveness of the others. Policymakers should therefore focus on identifying the ways in which the systems interact, attempting to facilitate positive interactions and minimize negative ones. An example of the different systems working together effectively is when journalists use documents produced in law enforcement actions, package them into a compelling story, and diffuse the story widely, thereby generating a public reaction that holds the powerful to account.303
The problem is that big corporations are well aware of the interactions between the systems and can utilize their power to neutralize “positive” interactions and maximize “negative” ones. For example, by using their market power to force mandatory arbitration and class action waivers on counterparties, big corporations not only reduce the expected legal sanction but also neutralize the abovementioned positive spillovers from the legal to the reputational system of control. Similarly, big corporations can use their vast resources to effectively shut down the threat of damning information coming from litigation by paying individual claimants (or public enforcers) off: paying heftier settlements in exchange for confidentiality provisions.304
Wells Fargo’s phony-accounts scandal is illustrative. Investigative reporters at the L.A. Times were the ones breaking the story, detailing how pressures to meet performance yardsticks led to massive consumer fraud.305 What allowed the reporters to come out with the story were court documents from wage-and-hours lawsuits filed by disgruntled Wells Fargo employees. The court documents provided readymade, libel-proof testimonies by former employees, detailing the pervasive lawlessness. Following the Supreme Court’s 2018 Epic decision,306 such workplace disputes are less likely to be aired in court going forward. The next time a journalist attempts to expose shenanigans by giant companies, damning inside information is more likely to remain out of reach.
Even more worrying is when big corporations create negative interactions between the systems. For example, big corporations may use their clout over the regulator to dilute the reputational sanction. To concretize, chemical companies accused of mishandling toxic emissions often refer to the EPA’s choice not to ban or regulate the chemical as proof that nothing is wrong with how the company handled it.307 Regulated companies know well that regulators are anything but omnipotent, and that regulatory inaction does not mean endorsement. In fact, regulated companies often do all they can to make sure the regulator cannot diligently monitor the situation. They use one arm to delay regulatory action while waving the other arm in a “nothing to see here” fashion, assuring the public that a diligent regulator is on the case. The illusion of regulatory compliance distracts the market from recognizing warning signs of corporate misbehavior.308
A more proactive version is using the regulator as a validator who conveys the company’s version to the public. A classic example is companies writing the regulators’ press releases. The 2014 Pulitzer Prize–winning investigative report, for example, detailed the practice of energy companies sending drafts to attorney generals, who then tweak a few words, copy-paste, and release the draft on their official letterheads.309 DuPont’s intra-company emails provided a clear case in point. In one of them, the company’s VP insists that “the only voice that can cut through the negative stories, is the voice of EPA. We need EPA . . . to quickly (like first thing tomorrow) say the following: Consumer products sold under the Teflon brand are safe.”310 And indeed, as part of negotiating a settlement, DuPont insisted that the EPA include certain quotes in its press release, and then embedded those quotes extensively in the company’s own communications.311 Similar dynamics are at play in the context of SEC enforcement, whereby the SEC apparently lets big market actors negotiate the language of the press release that announces the enforcement action and settlement.312
Policymakers should strive to make the different systems interact with each other positively, such that one system’s advantages check and balance the other systems’ flaws. This necessitates looking at seemingly disparate areas such as civil procedure doctrines on approving protective orders and confidential settlements. To use our recurring case study, note that the C8 story became synonymous with DuPont, but C8 was produced for decades by another company, namely, 3M. The media coverage (and Hollywood, in the film Dark Waters) focused on DuPont, partially because of all the internal documents that were exposed in litigation against it in West Virginia. By contrast, the judge in the litigation against 3M in Minnesota sealed 500,000 3M documents back in 2005, effectively keeping that part of the story away from the public eye.313 A different policy on sealing documents could have changed the reputational sanction for polluting.
C. Evaluating Existing Policy Proposals
Armed with a better understanding of the features of bigness that make it harder to control corporate wrongdoing, we can reevaluate the desirability of three of the most common policy proposals: increase the severity of sanctions, ramp up regulatory enforcement budgets, and promote whistleblowing.
First, perhaps the most common proposal for fighting corporate wrongdoing is to increase the severity of sanctions, such as by sending high-level decision makers to jail, debarring and delisting recidivist companies, and so on.314 The analysis here suggests that such proposals are less likely to work in the context of super-large corporations.
Increasing the sanction does not address the two underlying issues of information and power. A robust finding of the empirical literature on deterrence is that higher sanctions rarely matter if the probability of detection remains low.315 Applied here, the threat of a bigger sanction down the line does not make it easier to prove culpability inside big corporations to begin with. It does not dispel enforcers’ fear of collateral consequences, but only aggravates it. Nor does it alleviate the disengagement and information silos dynamics, whereby super-large corporations do not comply with the law even when a rational calculation would suggest that they should. In fact, raising the sanction may backfire, as it heightens big corporations’ incentives to leverage their power to avoid detection and fight with all their might against enforcement.316 This, in turn, would make it even harder to investigate and establish liability in court, and would push enforcers to settle even earlier.
A second common proposal is to significantly increase the budgets of enforcement divisions so that they will be able to conduct complex investigations and build cases against top-level individuals.317 Here as well, our analysis casts doubt on the efficacy of such a proposal. Granted, regulators’ scarce resources present a significant hurdle for deterring misconduct in super-large corporations. Yet, increasing the budget would not solve other hurdles that this Article has highlighted, such as regulators’ reputation concerns. Regulators are incentivized to maximize observable yardsticks that their overseers in Congress and the general public can relate to, such as the number of cases brought and amount of fines collected.318 Even with increased budgets, regulators anticipate that vigorously pursuing cases against massive corporations would greatly reduce the number of overall cases that they can bring, and probably also the amount of fines collected (as they could more easily collect money by settling early). Further, regulators anticipate losing many of these hard-fought cases, which could make them appear worse in the court of public opinion. Fighting big corporations all the way to a verdict would therefore remain a losing proposition from a regulators’ reputational perspective.
Ideally, we would want to see a shift in regulators’ enforcement priorities, along the following two dimensions. First, regulators should dedicate more resources to pursuing big cases against the biggest market actors, even if it means compiling far fewer enforcement actions annually. Pursuing a few big cases would come with positive externalities in the form of better reputational deterrence and moral deterrence. The process in itself, regardless of the outcome, would inject quality information into the market on the behavior of the most important actors, and provide opportunities for judges and regulators to signal what is considered proper or improper behavior.319 Second, regulators should dedicate more resources to post-settlement monitoring, even if it means, again, opening fewer new cases.320 Without robust post-settlement monitoring, there is little hope of reducing big business’s recidivism.
Granted, such shifts in regulators’ priorities are unlikely to occur without a corresponding shift in how Congress, the media, and the public evaluate regulators. Regulators must feel free to focus on the process rather than on observable yardsticks. But starting in the last weeks of 2021 and into 2022, we received strong indications that such a shift is already in place, when Deputy Attorney General Lisa Monaco, FTC Chair Lina Khan, and DOJ Antitrust Head Jonathan Kanter announced their intentions to prioritize individual accountability and consequential cases with less emphasis on the immediate outcomes.321
A third common proposal by scholars who decry corporate misconduct is to promote whistleblowing, such as by instituting better protections against retaliation and more financial incentives for blowing the whistle.322 Recently, concrete legislative proposals were made along these lines in the United States323 and overseas.324 The analysis here reinforces the desirability of such proposals. We saw that it is harder for outsiders (regulators, journalists, NGO watchdogs) to detect misconduct within super-large, complex corporations. Credible damning information would therefore have to come from insiders. But we also saw that it is harder for insiders in these corporations to blow the whistle, given that retaliation is likely to be costlier: fewer viable employment opportunities to switch to, heavier retaliation by community members for going against the largest employer and donor in town, and so on. Accordingly, if we wish to prevent the next C8-like debacle, our best chances come from proactively encouraging people from within these mammoth-sized corporations to step forward and protecting them once they do.325
There is a broader point in play here. Unlike the first two proposals, promoting whistleblowing has better chances of success because it addresses one of the core problems with controlling bigness, namely, information. The next Section offers additional, less conventional ways to directly address the information and power problems.
D. Addressing the Information and Power Problems
Any attempt at solving bigness-control problems should address the perverse incentives of top decision makers to remain ignorant.326 How do we ensure greater involvement of top management in oversight? A key part of the solution could come from corporate law, with its director oversight liability doctrine.
A quick primer on these Caremark duties (named after Delaware’s seminal court decision) is in order. Caremark changed directors’ oversight duties from reactive to proactive: directors must install a system that monitors compliance issues and reports them back.327 On paper, such a doctrine mitigates the incentives of top-level executives to remain ignorant. But in reality, corporate law’s role in assuring compliance has traditionally remained very limited.328 The Caremark standard “was set high, effectively demanding that plaintiffs show scienter without having access to discovery.”329 As a result, derivative actions against directors for failure of oversight were routinely dismissed at the pleading stage. In recent years, however, there has been a shift: Delaware courts are now increasingly willing to apply enhanced oversight duties (under the “mission critical compliance” designation), and to provide outside shareholders with access to internal company documents in order to investigate potential failures of oversight.330
How do these reinvigorated Caremark duties apply to bigness? Historically, size served to dilute the prospect of Caremark liability. In the prior precedent of Allis Chalmers,331 the Delaware Supreme Court based its refusal to impose liability on the presumed inability to monitor misconduct in massive corporations. The court described in great length the bigness aspects of Allis Chalmers, from its sheer number of employees (over 30,000) and plants (twenty-four), to its inherent complexity and decentralized decision-making processes. In such super-large corporations, the court concluded, it is not practicable to expect directors to know when company employees engage in misconduct.332
Today’s version of Caremark liability, however, provides much stronger incentives for top managers in big corporations to proactively monitor wrongdoing.333 First, courts today increasingly employ the “mission critical compliance” designation, which means that instead of focusing solely on what directors actually knew, the courts are telling directors what they should have known regarding certain risks that go to the core of the company’s business. The mission critical compliance doctrine gives more weight to culpable ignorance, thereby fighting off incentives to remain ignorant.334 While initially appearing in the context of monoline, namely, smaller companies (those that only sell one product), today, the mission critical compliance doctrine is being applied to mammoth-sized corporations. In the August 2020 Chou case, for example, pharmaceutical giant AmerisourceBergen was embroiled in criminal activities regarding cancer drug repackaging.335 The violations occurred in one of the company’s many subsidiaries and involved only a tiny fraction of its overall revenues.336 Still, the court applied the mission critical compliance doctrine, implying that for some industries that operate under heavy regulatory scrutiny, many risks count as critical.337 An even better illustration came in the September 2021 Boeing case, where the court applied this enhanced scrutiny mode to directors of the aviation giant (a 140,000-employee company).338 Such decisions incentivize directors in massive corporations to constantly collect information and act on red flags in all parts of their firm.
Second, the courts’ increased willingness to let outside shareholders inspect the company’s books and records prior to filing a lawsuit puts pressure on corporations to document the upward flows of information. In Boeing, shareholders got access to over 44,100 internal documents related to how the company handled air safety issues.339 One could argue that such pre-filing discovery would only aggravate the incentives to not document. Yet the courts have clarified that any document not produced could be used as pleading-stage evidence against defendants.340 In the old mode of Caremark litigation, directors could point to lack of documentation as evidence that they were not aware of the problem, which in turn would let them off the hook (no bad faith).341 In the new Caremark era, by contrast, lack of documentation can serve as evidence of a lack of needed follow-up actions on the part of the board to remedy potential oversight issues.342 The emphasis on paper trails carries prophylactic value by mitigating the incentives to avoid information ex ante and by making management involvement more provable ex post.343
Promoting whistleblowing could therefore improve inside-out information flows, and recalibrating director oversight duties could improve bottom-up information flows inside large organizations. But we would still need to address the power problem, namely, the ability of super-large corporations to shape the environment they operate in. Here, one concrete solution would be to make the abovementioned mandatory arbitration and gag provisions unenforceable for big corporations with market power.344 As shown in Section V.B above, a public mode of dispute resolution could also generate positive externalities in the form of quality information on the behavior of large corporations, thereby boosting the ability of investigative reporters to hold big business accountable (interactions between the systems). Another, more classic solution to the power problem is antitrust. This is where we move to now.
E. Revisiting the Heated Antitrust and Corporate Purpose Debates
1. Antitrust
To the extent that bigness comes with control problems, stricter antitrust enforcement becomes more desirable, all else being equal. By reducing the size and complexity of big corporations, antitrust intervention could arguably have the added benefit of improving the ability of laws, markets, and morals to deter corporate misconduct.
However, I should be careful not to overstate my claim. We saw that the evidence does not suggest that big is categorically ungovernable; in fact, there are areas where it is more governable. Even if one concludes that bigness comes, on average, with control problems, there exist many other advantages and disadvantages to size besides governability. Whatever disadvantage size may create on the deterrence front would therefore have to be weighed against other factors. Further, one could claim that judges applying antitrust law to specific cases are ill-equipped to factor the elusive links between bigness and deterrence.345 Additionally, the analysis presented here does not suggest that competition is a panacea for deterrence problems. In fact, fierce competition may actually exacerbate some of the dynamics described here.346 Many of the cases we discussed were fueled by pressures to succeed in intense competitions with low margins.347 We should be wary of conflating the observation that monopolies behave badly with thinking that breaking up the monopoly would solve the bad behavior. The implications of this analysis to classic antitrust law are therefore limited.
With these caveats in mind, let us highlight three areas in which this Article can contribute to the current anti-monopoly debate and provide a new intellectual framework for thinking about when and how to breakup or downsize super-large corporations. First and most basically, this Article adds to our understanding of the noneconomic effects of size and concentration. Today’s “neo-Brandeisians” claim that concentration is bad regardless of its effects on consumer prices. The analysis here fleshed out how exactly concentration may create non-consumer harms, such as environmental degradation.
Second and relatedly, the analysis here puts a thumb on the scale of structural reforms instead of conduct rules.348 Critics of the anti-monopoly movement often acknowledge negative sides to big business but argue that these can be readily ameliorated with prudent regulation and countervailing powers.349 The analysis here cautions that regulatory regimes that look good on paper are too often malleable to the power and fragmented knowledge dynamics of massive corporations. Ordering these corporations to stop defeating emission inspections or to stop monetizing customers’ private data may not be as effective as it is in normal-sized firms.350
Third, this Article provides blueprints for tailored interventions that go beyond traditional antitrust tools,351 such as recalibrating director oversight duties and making mandatory arbitration and gag provisions unenforceable. In this way, this Article answers recent calls of anti-monopolists to (1) shift focus from criticizing the status quo (“big is bad!”) to providing concrete analytical frameworks and policy proposals;352 and (2) expand the focus from protecting consumers qua consumers to protecting workers, communities, the environment, and other aspects that can be exploited through consumerism.353
2. Corporate Purpose
Corporate purpose has reemerged in recent years as the hottest topic in corporate governance.354 The deep unease with corporate behavior was not lost on corporate America, and several key market arbiters reacted by reversing course on shareholder primacy and suggesting that companies should focus on broader societal objectives.355 This notion has since found its ways to concrete legislative proposals.356
Importantly here, the corporate purpose debate has always been about bigness.357 The topic initially emerged during the 1920s, against a vision of an “American economy dominated by a small number of gigantic, stable corporations that essentially control the nation’s business.”358 To this day, “[w]hen legal commentators discuss corporate social responsibility, they really mean the social responsibility of giant corporations.”359 Accordingly, the insights developed here on bigness can inform the debate.
To illustrate how the analysis here corresponds with the corporate purpose debate, we need look no further than Milton Friedman’s famous maxim, namely, that the only social responsibility of business is to increase its profits.360 Friedman based his maxim on several assumptions that are pertinent here. First, he suggested that corporations “make as much money as possible,” but only “while conforming to the basic rules of the society, both those embodied in law and those embodied in ethical custom.”361 Second, he said that a singular focus on maximizing profits only makes sense when corporations operate within “open and free competition without deception and fraud.”362 If these assumptions hold, managers who focus on maximizing profits will improve not only their own company’s bottom line, but also overall welfare, while other bodies of law regulate corporate externalities. Yet if this Article has shown us anything, it is that these assumptions often break down when dealing with super-large corporations.
Big corporations are neither “price takers” (because they have market power) nor “rule takers” (because they have political power and can dilute the expected sanctions). They do not take “laws” or “ethical customs” as given, but rather influence them to suit their purposes.363 As a result, Friedman’s maxim, even according to his own framework, does not apply to them. A strict focus on shareholder wealth maximization in these companies can prove bad for overall welfare. From an economic point of view, companies with market power can maximize shareholder wealth by reducing output and increasing prices.364 From a governability point of view, companies with political power to influence the rules can maximize shareholder wealth by externalizing greater costs on others in society.365
We thus cannot think of corporate governance in our largest corporations as strictly a concern for shareholders. Ideally, we would want such companies to develop organizational cultures that take their effects on society at large into account.366 However, we cannot rely on Business Roundtable–like voluntary declarations by big corporations that they have seen the light and are now going to take care of broader societal interests. Such declarations often amount to cheap talk at best or co-opt regulation and entrench management at worst.367 Relying on “the business case” approach to rein in social bads would only make big corporations focus on the most salient social issues. They could continue to be leaders in sexual harassment prevention or board diversity, which are noble causes, while ignoring other worthy causes with less saliency and more complexity.368
As is in many other fields, “self-regulation” here would only work effectively if there were credible threats of legal intervention, media scrutiny, and market pressures.369
Conclusion
“[T]he large and growing [societal] harms from corporate wrongdoing have become one of the [most] seminal issues of our time.”370 In order to better deter such corporate wrongdoing, we first need to understand its antecedents: the factors that are associated with more or less wrongdoing. This Article has highlighted bigness as one such factor. By weaving together research from multiple disciplines—social psychology, communications, economics, behavioral ethics, political science, and more—this Article spotlighted how firm size and market power affect governability. This Article also provided recommendations for policymakers on how to address the problems and for academics on how to continue thinking about them.
Existing accounts of the problems with deterring corporate wrongdoing tend to be single-institutional: they focus, for example, on how settled concepts in criminal law make it ill-equipped to deal with modern mammoth-sized corporations, or how regulators are captured by concentrated industries with a few big actors. This Article, by contrast, showcased the need to think about governability holistically: corporate deterrence comes not just from laws, but also from reputation concerns, moral constraints, and the interactions between these different systems. Thinking about corporate deterrence holistically allows us to identify the areas that are more worrisome and those that are more fixable from a societal perspective.
This Article also has notable limitations. Figuring out the conditions under which big corporations are more likely to subvert rules is an extremely broad issue that does not lend itself to clear empirical proofs or neat models. Lab experiments meant to test how individuals behave cannot capture the bigness element, namely, how operating within big corporations with their complexity, authority relations, and fragmented knowledge affects the propensity of individuals to misbehave.371 Yet neither the fuzzy nature of this topic nor the nascent state of our understanding of it should deter us from delving into it further, if only because the stakes are extremely high. It is not just that the costs of misconduct by big corporations are in the trillions, but also that concrete policy proposals to break them up or intensely regulate them are constantly being raised these days. And while we could not find easy answers to the broad “is big ungovernable” question, we gained, in the process of addressing it, a better understanding of the hitherto understudied special challenges that bigness presents to accountability.
We also opened up space for thinking further about bigness in various legal fields: from antitrust, to civil procedure (secrecy versus openness), to corporate law (director oversight duties), to areas we have not developed here for considerations of scope, such as international private law and regulation.372 Beyond this or that specific policy proposal, a general lesson emerging from our discussion is the need to think about tailoring legal institutions differently as a function of size. In that respect, our discussion corresponds with several recent proposals, such as Georgiev’s proposal to rethink securities laws’ disclosure requirements,373 or Diamantis’s proposal to adopt a sliding test to criminal law doctrines of knowledge that account for the corporation’s size and complexity.374
The potential for contributions that were not developed here only strengthens the message that much work remains for legal scholars in understanding how bigness affects the institutions we too often take for granted. Bigness is now being invoked daily in presidential speeches, major media outlets, and congressional discussions, yet the public debate tends toward big polemics and strong sentiments.375 This Article represents a step toward injecting much-needed theory and evidence into the discussion.