Defining corruption as the exercise of public power for private, selfish ends, many theorists have argued that individuals can be corrupt even if their actions are legal. This essay explores the knotty question of when legal corporate action is corrupt. It argues that when corporations exercise public power, either through monopolistic control of a market or through campaign contributions and support of governmental actors, they are subject to the same responsibilities of anyone who exercises public power. Therefore, as a theoretical matter, we should call corporations corrupt when they exercise public power selfishly, in a way that puts their own interests over the public's interests. Because they make legal corporate corruption less likely, global anticorruption campaigns should therefore emphasize antimonopoly laws and campaign finance laws.
Should we call legal corporate political behavior corrupt? If so, when?
It is a tricky issue. Of course, in some cases, corporate actors engage in illegal bribes of public officials, and we can easily label this behavior corrupt.1 But more frequently, corporate actors use sophisticated legal means to exercise power over public officials: by making campaign contributions, lobbying, exerting media influence, funding nonprofits, sponsoring think tanks, paying speaking fees, or even cornering the market on key goods and services, creating public dependencies on the corporation. These kinds of behaviors make up what Michael Johnston has termed “influence markets,” which he identifies as the primary mode of corruption in developed democracies.2 These behaviors are also explored in depth in the works of sociologist Amitai Etzioni.3 All of these behaviors are not only legal in the United States, but are encouraged and taught as essential strategies in business schools. They also have the tendency to spread. Having built their power within the United States or similar legal systems, corporations then use legal tools to exert influence in other countries. Depending on which side of the law they stand on, corporate actors may push to legalize the most powerful of their mechanisms of control, criminalizing the tools used by weaker societal agents, or they may exercise their influence to decriminalize their behaviors in a new market. The question is, which of these behaviors should we call corrupt, and which are merely corrupting?
In 1820 America, it was not illegal for a corporation to give money to a member of Congress in explicit exchange for that congressperson's vote. In 2017 America, because of Citizens United, it is not illegal for a corporation to spend millions of dollars to punish a congressperson who voted against its interests. We can certainly agree that the former is corrupt; I think most would accept that the latter is also corrupt. But if legality is not the line between corrupt and noncorrupt corporate political behavior, what is?
I argue that we should use the same test for corporations as we do for public officials, condemning selfish behavior as corrupt when it accompanies the exercise of public power, regardless of whether that public power derives from formal office-holding. Elected officials who exercise public power in the service of private ends are corrupt irrespective of the legality of their behavior. By extension–with understanding that it is not easy to identify what constitutes “public power” or even “selfish behavior”–all selfish exercise of public power is corrupt. The key theoretical point is this: public power, not public office-holding, ought to be our marker for determining who may be guilty of public corruption. Corporate actors are corrupt when they exercise public power in a way that serves selfish ends at the expense of public ends, regardless of whether it is illegal, and regardless of whether they formally hold office.
The descriptive implications of this conclusion are substantial: it means that some of the great drivers of contemporary corruption around the world today are large multinational corporations engaging in legal behavior. The practical implications are also substantial, and flow from the improved description: our anticorruption strategies must include antimonopoly laws, not because antitrust violations are themselves corrupt or because mergers are themselves corrupt, but because corruption is more likely when economic power is centralized. Failure to name legal corporate behavior as public corruption in global anticorruption campaigns to date has led to a focus on passing criminal laws and transparency laws, instead of examining problems of market structure and monopolization with global and domestic impacts. As Lord Acton famously put it: “Power tends to corrupt.” Power is especially likely to corrupt when it is unconstrained by democratic accountability.
This kind of corporate and multinational corruption is a tragedy of design. It flows from our failure to protect markets from concentrated economic power. Corporate monopolies are a result of legal frameworks that enable excessive concentration of private power, limit the freedom to engage in moral action by officers and directors, and create overwhelming incentives to bend public power to selfish ends. Unlike small companies that have limited incentive or capacity to corrupt–because they do not exercise public power–multinational corporations, at a certain size and with enough power, are built to corrupt.
The critical strategic solution to this design flaw is to engage antimonopoly laws in anticorruption efforts. The antimonopoly approach is prophylactic instead of punitive; in this way it resembles elections, another prophylactic anticorruption tool. New antitrust enforcement should not seek to punish corrupt behavior, but to encourage structures of power that make corruption less likely. Open markets, free from dominant players, are not only important for a thriving economy and innovation, but for limiting corruption.
This essay proceeds in two sections. The first makes a theoretical argument and shows that a surprising formalism pervades many approaches to understanding public corruption. This formalism appears in two ways: First, discussion of public power often stops with a formal analysis of who holds a particular office, instead of who wields power over that office. Second, even those anticorruption analysts and activists who claim not to tie a definition of corruption to legality tend to use legality as an important marker in separating the corrupt from the noncorrupt. Building on these theoretical points, the second section highlights antimonopoly and campaign contribution laws as critical sets of tools for dealing with this crisis of corruption.
To be clear, I do not make accusations about corrupt behavior by particular modern corporate multinationals. An approach of identifying after-the-fact bad actors is always going to be a weak strategy. Instead, I lay out a theoretical framework for enabling accusations against modern corrupt corporations and a practical road map for deterring future corruption via structural changes.
Aristotle laid out six kinds of government: three ideal forms and three corrupt forms. The rule of one he described as either monarchy or tyranny; the rule of a few as either an aristocracy or an oligarchy; and the rule of the many as either a polity or mob rule. The fundamental difference between the good and corrupted government, according to Aristotle, was the psychological orientation of those who governed: corrupt governments were selfish; ideal governments sought the public good. Explaining the difference between a tyranny and a monarchy, he wrote, “the tyrant looks to his own advantage, the king to that of his subjects.” A tyrant is a king who “pursues his own good”; an oligarchy is an aristocracy that pursues its own good; and mob rule is a publicly governed polity whose constituent parts each pursue their own selfish interests.4
This framework, which I have adopted, suggests there are two key features of corruption: the exercise of governing power and selfish intent. The implication of this framework is that private actors engage in public corruption when they wield governing power selfishly.
Within the anticorruption field, there are those who describe corruption in terms of the violation of formal roles and obligations, and those who see corruption in terms of the illegitimate pursuit of private interest at the expense of the public interest.5 The former ties itself in knots of positivism. As political scientist Richard Mulgan has recently argued:
By taking existing duties and rules as given, such definitions are too closely tied to a particular institutional context. They do not provide an external standard by which to assess whether the duties or rules themselves prohibit actions that should be regarded as corrupt.6
All parties appear to agree that public power is an important feature of public corruption, though this has been given short shrift in some of the literature.7 For instance, a recent article appearing in the UCLA Law Review observes that most definitions of corruption involve the abuse of public office for private gain. The article continues: “The term ‘public office’ is relatively clear. It includes, among others, those persons whom the electorate has entrusted with power to advance the public interest.”8 Accompanying this assertion is a link to a judicial decision about the scope of a bribery statute, making the easy error of conflating statutory law and definitions of corruption in one area but not in another. If one does not confine oneself to statutes, it is not at all clear that office, instead of power, is the key question.9
For much of industrial history, private parties were viewed as corrupt when they exercised public power, regardless of whether they held office.10 In the 1874 case Trist v. Child, an old man hired a lobbyist to help collect a debt from Congress. After the lobbyist succeeded, the old man refused to pay him; in response, the lobbyist sued the man for money owed. The case came before the Supreme Court, which had to decide whether contracts to lobby were legitimate and enforceable in court. The Court concluded that they were not, writing that “If any of the great corporations of the country were to hire adventurers who make market of themselves in this way [for] the promotion of their private interests, the moral sense of every right-minded man would instinctively denounce the employer and employed as steeped in corruption.”11 The Court's language indicates that corporations could themselves be corrupt, not merely a means by which public entities are corrupted.
But over the last forty years of anticorruption efforts, many academics and journalists have treated private companies as corrupt only when engaged in what is sometimes called “private corruption”: namely, accepting internal bribes or kickbacks.12 Much of the discussion about private entities–big multinational companies like Monsanto, Google, or Siemens–concerns whether we should recognize a category of private-to-private corruption.13 To address these concerns, some definitions of corruption focus on “entrusted power” instead of public power. Transparency International, for instance, defines corruption as “the abuse of entrusted power for private gain,” in order to include private-to-private relationships within the definition. But Transparency International does not have a clear scope of what constitutes entrusted power for purposes of public corruption, nor does it examine whether multinational corporations can be seen as having “entrusted power” because of their enabling statutes. Alternatively, private companies are seen as corrupting when they induce behavior on the part of elected officials, or perhaps when they break existing anticorruption laws. They are not treated as corrupt for their use of legal mechanisms, even when that use is for self-serving ends.
Some modern definitions openly rely on public office, instead of public power, as a central feature of corruption. Political scientist Joseph Nye's influential definition of corruption begins with a claim about the centrality of formal roles, arguing that corruption is either rule violation or “behavior which deviates from the formal duties of a public role because of private-regarding (personal, close family, private clique) pecuniary or state gains.”14 Several other scholars have placed public office at the center of the definition, but even those who do not privilege the phrase “public office” or “formal duties” often implicitly limit the accusation of public corruption to those with formal public power.15
How should we approach this question? It is perhaps easiest to divide the possible approaches into a formal approach and a functional approach. The formal approach limits the accusation of public corruption to those who hold an official position. The functional approach looks at whether public power is exercised, regardless of office-holding.
A formal approach leads to peculiar results. Imagine a rich business owner in a small town. He consciously chooses to use his wealth to elect a town council and mayor that will serve his interest and lower his taxes. He is shameless about his desires: he readily announces that he is only interested in himself, and will use whatever means he can to serve himself. A formal approach would treat the business owner as not corrupt because he is not an elected official. It would not ask how he uses his wealth, whether in fact he has public power, or whether he is being selfish. Instead, it looks merely at his status: since he was neither appointed nor elected, it treats him as someone who might be involved in private corruption (accepting bribes in his business) or someone who might lead to the corruption of public officers, but not someone who might be corrupt in his own right.
On the other hand, a functional analysis would treat this business owner as engaged in public corruption because he is using public power, and using it to serve private ends without regard to the public good. That he may also be corrupting the local government is a secondary question. This business owner is not unlike Aristotle's king (or oligarch), who chooses to rule over others in a way that benefits himself. That he uses the mechanisms of democracy does not change the fundamental combination of his ruling others and his moral orientation.16
Another thought experiment in formalism also leads to the mangling of language. Imagine a king who has inherited absolute power over his country. He is selfish and cares only for his own interests, not the interests of the public. Because he is worried about revolt, he chooses to install an elected government, but creates laws allowing for only one party on the ballot, and establishes informal mechanisms that ensure that he is the only person who can select who runs for office. He then officially steps down from his position and abolishes the monarchy. But there is no doubt that he controls who gets “elected” and what decisions they make in office. A formal approach would say that only those elected officials can be guilty of public corruption. A functional approach would consider the actual power dynamics, not the form.
As these examples show, a functional analysis is the more natural approach: formalism seems to simplify the concept, but adds a requirement to public power of public office-holding that is hard to justify. Ruling is what creates moral obligations, regardless of how that rule is exercised. The strongest argument against the formal approach is that there is no a priori reason to limit the scope of public corruption to those holding elected, appointed, or inherited office. The selfish interests in a corrupt government might be the interests of the people holding formal power in the government, but–critically–they can also be the private interests of someone or something that exercises informal power over the official government from outside it.
The best defense of a formal approach is that it is more administrable and renders corruption easier to measure. But we should not confuse the administrability of criminal and civil laws with the correct definition of a nonlegal term like corruption, just as we should not refuse to call something corrupt because it is difficult to measure. The functional approach would be inappropriate for defining criminal laws of corruption; it would require a fact-finder to make determinations of influence and power in a political society, beyond a reasonable doubt.17 But we are not rewriting legal definitions, and inadequacy in criminal law does not make the functional approach inadequate in our efforts to locate corruption.
Another possible objection to the functional approach might be that it seems harsh: it subjects private actors who have never run for public office or sought to be appointed to public office to accusations of public corruption and obligations to the public good that they never wanted. But on the individual level, this problem does not exist. Individuals are not required to exercise public power, even when they have the capacity to do so. And most ceos of most companies, like individuals, simply have no capacity to exercise public power. They are free to suggest ideas, set up meetings, and occasionally lobby officials, but no one would argue that in so doing they are exercising public power. Success in private business creates no obligation to engage in the public sphere in a selfish way. Moreover, inasmuch as those with inherited public power never chose their position, we do not soften the blow of corruption accusations by arguing that kings cannot be guilty of misusing powers they did not seek. They may always abdicate. However, for corporate officers and directors of enormous companies that can exercise governing power, this harshness does expose a fundamental problem with our current antimonopoly laws by creating two obligations that conflict with each other.
Using the functionalist approach, we should shift from an analysis of office-holding to an analysis of who holds “governing power.” Governing power exists when a company, person, or institution has the capacity to make choices that govern the lives of others. A juror has governing power over the defendant. A magnate has governing power over his town when he uses his ability to elect or defeat candidates who then exercise formal power. Governance is often defined by reference to a combination of decision-making and the implementation of those decisions. Political scientist Stephen Bell's popular definition of governance argues that it is “the use of institutions, structures of authority and even collaboration to allocate resources and coordinate or control activity in society or the economy.”18 The lines are by no means clear; and there is not space here to explore in full the difficult questions of what is and is not governing power.
More important, the job of anticorruption activists is largely not to identify instances of normative failure, but to identify the syndrome, and then push for the rules that make the syndrome less likely. We need not spend much time debating the particulars of who or what company is corrupt, so long as we agree that there is a broad set of powerful companies that pursue selfish interests while exercising public power.
By way of analogy, consider a national campaign against alcoholism. One way to deal with alcoholism is to try to identify everyone who is alcoholic–engaging in extensive studies to determine who might be dependent on alcohol and who is merely drinking a lot–in order to provide individualized resources to those who need them. In that approach, the question of who makes the judgment about particular individuals, and by what criteria they are judged, is critical. But another approach might be to use countrywide surveys to identify that there is problem of alcoholism, and then suggest countrywide solutions that would reduce the levels of addiction overall and the likelihood of future addiction. In the second approach, we spend little energy parsing the alcoholic from the nonalcoholic, and most of our energy is focused on prophylactic rules.19
Using this syndrome approach, undoubtedly there are several big multinationals engaged in public corruption. We need not have a consensus around individual actors’ corruption in order to agree that there is endemic corruption. Two analysts might disagree over whether Siemens or Amazon has more governing power, but they can certainly agree that some large multinational corporations engage in the selfish use of public power, and would likely include both Amazon and Siemens in that category.
For instance, I can argue that Google's exercise of public power is corrupt because it does so in pursuit of its own selfish ends, regardless of the impact on the public good. As evidence to support my argument, I could point out that, as of 2017, Google is the largest lobbyist in the United States. Google has been implementing a successful political strategy to embed its software in public schools (both in order to get its tools adopted and in order to collect data).20 Google is a major funder of think tanks and has exercised its funding power to shape policy, supporting scholars who support its own political ends. In short, an essential, nonaccidental part of Google's business strategy is to shape public policy in a way that serves its own narrow interests. As with the rich businessman controlling the small town described above, I argue that a functional analysis would treat Google as corrupt. However, one need not agree with my particular argument about Google in order to agree that the structure of power in our society makes it likely that powerful companies like Google–if not Google itself–will use public power for private ends.
The legality of the behavior is not decisive in determining either whether there is governing power or whether it is selfish. Google's practices as described here are entirely legal under U.S. law. Lobbying is legal, funding think tanks is legal, building a political strategy to shape public education is legal, and supporting academics is legal. Moreover, these behaviors should be legal. However, the legality or illegality of a behavior is not a particularly useful distinction in determining whether something is corrupt. As political scientist Dennis Thompson has argued, “Connections that are proximate and explicit, elements required to show bribery, are not necessarily any more corrupt than connections that are indirect and implicit. The former may be more detectable, but are not necessarily the more deliberate or damaging form of corruption.”21
Instead, there are many possible relationships between the legality of a behavior and its corruptness. First, it is possible that there is no relationship between illegality and corruption. The second possibility is that illegality separates corrupt from noncorrupt behavior. The third possibility is that illegal activity defines the heart of corruption, that which is easiest to define and which we should most readily condemn, but some legal activity is also corrupt, if less intensely so. The fourth possibility is that there is often a relationship between illegality and corruption, but that such correlation does not help us decide in any particular instance whether an action is corrupt or not. Corruption encompasses a great deal of legal behavior; only a small subset of corrupt behavior has been criminalized. Moreover, noncorrupt behavior can be criminalized and called “corrupt” by the state.
The final option–a correlative relationship but not sufficiently strong to make presumptions–is the best way to understand the connection between corruption and legality. While the overlap between illegality and corruption exists, and may not be wholly arbitrary, it approaches arbitrariness because the reasons for not criminalizing behavior are so varied and historically and culturally dependent. Unless one is a positivist (believing law defines morality), there is no a priori reason to assume a strong relationship between that which has been criminalized and that which is corrupt. In fact, given that power tends to protect itself, in most polities we should often start with the assumption that the most corrupt acts are shielded from criminal liability by those in power. Those in power, be they judges or lawmakers, may have selfish reasons to protect corrupt behavior and criminalize noncorrupt behavior. World history is littered with regimes that do not criminalize corrupt behavior because those in power are engaged in it. But even in a perfectly functioning democracy, where an engaged public would have criminalized corrupt behavior, there are many reasons for using other tools than criminal law to deter corruption.
The reasons for criminalizing some behavior and not others is often unrelated to the morality of the action or the degree of public condemnation. A democratic society could decide that criminal law is not a particularly effective mechanism for deterring corruption. In the United States, bribery of members of Congress was not illegal at the federal level until 1853. Before then, everyone thought that paying a congressperson in exchange for changing a vote was corrupt; they simply did not use criminal law as the tool for deterring such corruption. Other considerations, like the desire to protect certain forms of expression, could lead to the legalization of corrupt behavior. For instance, under existing U.S. law, a senator who accepts a personal gift of $15 with the understanding that it will influence his or her vote is committing federal bribery. No explicit exchange is needed. However, if the same senator accepts a campaign contribution of $5,000, knowing it represents the purchase of the exercise of one hundred votes, that does not violate federal bribery law in the absence of an explicit contract or agreement indicating the senator's intent.22 There are reasons, both historical and protective of political expression, that make the former a crime and the latter not. But those reasons tell us nothing about the corruptness of the action. The fact that the latter is not a crime is not evidence that it is not corrupt, or that it is somehow less corrupt.
As of 2016, it is legal under federal law in the United States for someone to pay tens of thousands of dollars to a state governor in exchange for the governor, using the official title of the office, setting up meetings and making introductions to other officials and business executives. The Supreme Court struck down a law criminalizing this behavior because of free speech and due process concerns. Nothing in the decision suggested that the Court thought that the behavior was not corrupt.23 In the same vein, lobbying, which was criminal behavior for one-third of American history, has achieved protected legal status because laws against lobbying were struck down as violative of the First Amendment.24 This is undoubtedly a good thing. But the fact that criminalizing a behavior would threaten free speech is hardly sufficient to mean that no instances of that behavior are exhibitions of corruption.
In sum, criminality and corruption may have a substantial overlap in certain developed democracies, but that overlap does not tell us much about the corruptness of any particular act, or whether most corrupt acts are crimes.
You might argue that I have created a straw man. It is the rare definition of corruption that openly relies on criminal law as a starting point for determining whether corruption exists. Definitions are far more likely to refer to “abuse of public power for private ends,” or “norm violation in a self-serving way by those in public power.” Even Joseph Nye, whose definition is often characterized as requiring illegality, recognizes norm violation, apart from illegality, within the category of corruption. However, among the scholars and commentators who theoretically acknowledge that much corrupt behavior is legal, many still exhibit an assumption that legality is a good marker of corruption. Empirical studies and economic models of corruption often start with criminality.25 Transparency International starts with the assumption that most corruption is illegal.26 Many comparative studies rely on criminality directly or indirectly.
For instance, in Susan Rose-Ackerman's landmark book Corruption and Government: Causes, Consequences, and Reform, she acknowledges that legal corruption is important, but states that because her work is comparative, she will only look at those instances in which laws were broken.27 She further argues that it “may be rhetorically valuable” to call legal behavior corruption, but that it does not “further the analytical” or “policy exercise of understanding the landscape and proposing reforms.”28 This seems to get the analytical and policy project upside down: it privileges those with the power to make the law with the power to define corruption.29 Rose-Ackerman's recent work has been more likely to recognize legal behavior as a significant problem, but I use this example because it is typical of the simultaneous acceptance and rejection of legal corruption.30
Once the anticorruption community accepts that neither office-holding nor legality is a definitive marker of the existence of public corruption, it becomes free to explore corruption as it actually exists in modern society.
In the last thirty years, the entire machinery of modern multinational corporations has developed, through law and culture, to embrace the pursuit of public power as an essential business function. The deep design of a large multinational corporation is to build power to gain control over local governments and international regimes in which it operates so that it can advance policies that create value for the corporation.31 Large multinational corporations routinely exercise public power, and do so guided by private interests above public ones. The intent/orientation of large corporations is easier to divine than the intent of most individuals or organizations. When there is a conflict between public and private interests, the enabling statutes of a corporation require an orientation toward a limited set of stakeholders.
One might point out that corporate entities need not seek short-term profits. As the U.S. Supreme Court recently reaffirmed: “Modern corporate law does not require for-profit corporations to pursue profit at the expense of everything else, and many do not.”32 The myth that corporations are required to maximize shareholder value is just that: a myth, and one that is largely pushed by activist hedge fund managers seeking to pressure corporations to produce short-term profits.33
However, corporations are not free to pursue the public good when doing so conflicts with the long-term sustainability of the corporation. Under state law, directors and officers of a corporation have a duty of care and of loyalty to the corporation. That duty does not flow merely to shareholders, but to all the stakeholders in a corporation. At a basic, ethical level, these laws create an obligation to maximize value–arguably long-term, sustainable value–for the corporation. It is rare that a lawsuit succeeds on the grounds that directors and officers violated these obligations, but that does not mean that the obligation does not exist. Instead, the laws, designed to ensure that directors and officers do not treat the corporation as their own vehicle, also ensure that the public good cannot justify decisions that directly hurt stakeholders.
In many instances, corporate stakeholder ends will either support the public interest or at least be consistent with the public interest. In these instances, there is no corruption problem. Under other circumstances, a ceo may have some discretion due to conflicting visions of long-term sustainability: this discretionary space is where corporate social responsibility (csr; a form of corporate self-regulation) is likely to be most powerful. In the overwhelming majority of other instances, the corporation will not exercise public power. In these cases also, there is no corruption problem. For the millions of small or medium-sized corporations, their private obligations will not conflict with public obligations, because such corporations simply do not have the power to shape public policy on taxes, trade, antimonopoly, or contracting: they will face no moral dilemma. The local pizza shop has no raft of lobbyists, and if the owner makes a political donation, it will be $30 or $300, not a $300,000 independent expenditure. The vast majority of companies never engage a lobbying firm, let alone build all the tentacles of public-policy-bending machinery. Some company owners may be wonderful and deeply invested in their community, others greedy and self-centered, but as a structural matter, these companies do not pose a public corruption threat.
However, for large corporations that invest heavily in politics, there will be frequent episodes in which the obligations to long-term profitability and to the public interest directly conflict. Four of the most common conflicts involve tax laws, trade laws, antimonopoly laws, and contracts with the government. Big corporations will almost always have an interest in lowering their tax burden, improving their position in global trade, decreasing antimonopoly enforcement, and increasing opportunities to win government contracts. Occasionally these interests will align with those of the public, but frequently they will not. It is indefensible–to all the stakeholders in the corporation–not to be engaged in politics and not to build public power that can be used to benefit the corporation in terms of taxes, trade, antimonopoly, and government contracts. A ceo of Apple that did not have a public relations firm would be fired by its board of directors.
Imagine a ceo of a modern multinational corporation with $100 million to invest. She can choose to invest the money in decreasing the cost of producing the product, or she can invest the money in changing the laws to decrease the corporate tax rate. The first involves changing the production line, switching some materials, and a slight product innovation; the second involves a combination of campaign contributions, direct lobbying, media strategy, and coauthored white papers. Most estimates suggest the first strategy provides a 5 percent return on investment, while the second strategy provides a 50 percent return on investment. The first strategy does not hurt the public at large; the second strategy decreases essential tax revenue for schools. The first strategy involves no corruption. The second strategy is corrupt. We would expect the ceo to engage in the second strategy. The selfish exercise of public power–public corruption–is an essential part of the job.
How can we change that behavior? How can we fight the threat of rampant legal public corruption by large multinationals? Some analysts, like Ben Heinemann Jr., argue that the discretion afforded directors and officers is far greater than that which they exercise, and that corporate leaders can, consistent with law and culture, pursue the public good. Heinemann's efforts are important, but cannot address the problem posed by a corporation like Apple that wants to reduce its tax burden through lobbying and campaign contributions. Some might argue for a fundamental overhaul in corporate law, explicitly requiring officers and directors to serve the public good. And the rise of new corporations operating with clear public obligations might create positive impacts at the margins, but the side effects of fundamentally restructuring the corporate form would be far from benign. Moreover, this argument is antidemocratic, and essentially an argument for aristocratic/oligarchic rule: it accepts that multinationals play a governing role, and merely requests that they do so with a public orientation.
Instead, our anticorruption efforts should focus on the precise point at which public corruption comes into play: when corporations come to exercise public power. Corporate public corruption is most likely when the industry itself is very large and heavily concentrated; when there are cross-industry interests in bending public power; or when a single corporation has become essential to a polity, or “too big to fail.”
In other words, we should focus public policy on the problem of corporations exercising public power–which only happens at a certain scale and degree of power –and not the problems of corporations being selfish: let them be selfish, but do not let them govern.
In the United States, there is a long tradition of resisting the corrupting tendencies of concentrated power through antimonopoly laws. These laws–at the center of which is the Sherman Antitrust Act of 1890–were not designed to punish corrupt behavior, but to make corruption less likely. They were designed to prevent corporate directors and officers from facing the point at which their public and corporate obligations clashed. They were designed to ensure that private parties did not gain unaccountable public power. As Justice William O. Douglas explained in his dissent in the 1948 Supreme Court case U.S. v. Columbia Steel Co., the traditional philosophy of American antitrust law is that
all power tends to develop into a government in itself. Power that controls the economy … should be scattered into many hands so that the fortunes of the people will not be dependent on the whim or caprice, the political prejudices, the emotional stability of a few self-appointed men.
Drawing on that tradition, we should embrace antimonopoly law as an essential tool for fighting local and global corruption.
The first target might be highly concentrated industries. Profits are higher in concentrated industries, creating more cash flow for investment in politics. (It is no accident that pharmaceuticals, an industry that explicitly relies on monopolies, has among the highest profits and the greatest political investments.) Moreover, it is simpler and cheaper to organize a group consensus when the potential members are few. Fewer actors can more easily make joint strategic decisions about what to demand from government and create a shared, consistent message when lobbying and in meetings. With fewer actors, the costs of identifying shared needs, of coordinating timing, and of identifying and punishing free-riders are all reduced. In monopolistic or oligopolistic industries, it is easier to share fixed costs, like writing legislation, identifying targeted politicians, and producing effective messaging. The concentrated industry therefore can more economically lobby for shared goals, including decreasing taxes for the industry, increasing subsidies for the industry, decreasing regulations, and creating public insurance for the industry. An essential part of our anticorruption strategy, then, must be decreasing concentration. That means looking at industries that are dominated by few firms, such as online advertising or online retail.
Anticorruption reformers should also focus on corporations that have grown so large that they represent a significant fraction of the economy. When the size of a corporation relative to the gdp is significant–like 2 percent of gdp–democratic choices become constrained by the self-interest of the individual corporation. Even in the absence of resources devoted to purchasing political influence, the company with a large relative size will have public power. Its sheer size makes it incumbent upon legislators to design laws that will at minimum ensure the stability of the company. If Lockheed goes under and lays off all of its employees, that has an impact on the entire economy. Even without lobbying, therefore, Lockheed can make demands of government based on the threat of its own failure. Companies that are large relative to the size of a country's gdp can control politics by threatening to collapse or leave if their demands are not met.
In concrete terms, global anticorruption should support free and open markets, with decentralized economic actors. We should support antitrust efforts that put barriers in the way of companies’ monopolistic behavior, such as the European Union's efforts against Google; support antitrust regimes that lead to breaking up heavily concentrated industries and stopping mergers; and support campaign finance regimes that make corporate influence on elections more difficult. The goal is to encourage an approach toward power that recalls Justice Louis Brandeis: concentrated private power is corrupt and corrupts, and therefore should not be allowed.
The most useful antimonopoly, anticorruption strategies will differ in particular contexts; but as the exercise of power by multinationals continues to grow, there are a handful of urgent approaches:
Applying neutrality principles to platforms like Google and Facebook, and not allowing vertical integration: search services and advertising must be broken up. Amazon and Apple must be limited in their ability to discriminate in price or search, and to use pay-to-play models in their search. The massive public power and control wielded by these platforms depend on their ability to leverage their power in one area to make profits in another. This approach includes condemning countries that refuse to limit platform dominance and power.
Supporting legal regimes that separate distribution from content in cable and wireless companies, requiring the breakup of Comcast, for instance. Condemning countries that refuse to separate the two.
Urging countries to break up big banks, both in terms of size and function.
Supporting the breakup of the monopolies of companies like Monsanto, allowing for competition from farmers; opposing the Monsanto-Bayer merger; supporting countries that ban the ownership of seeds and chemicals.
Encouraging global trade agreements to disfavor monopolistic practices.
Condemning countries that allow corporate spending in elections.
None of these principles is simple to implement. There will necessarily be a high degree of over- and under-inclusiveness in any rule. There is no magic number representing company size within a country, or across countries, and no magic structural relationship that will avoid these harms. This, of course, is true for most laws: even for something seemingly more straightforward like traffic law, there is no magic number at which the speed limit best accommodates the principle of reducing unnecessary deaths. But when it comes to governance and rules of governance, there is always special difficulty in defining the rules of the game, because the rules of engagement create the outcome, including the outcome of what the rules of engagement should be. However, the difficulty in designing rules should not be a deterrence to trying. The underlying argument here is similar to that of the mid-twentieth-century Chicago school of economics. Our visions of human nature differ: I believe people are complicated and can be public-orientated, that we are not solely or even primarily homo economicus. And we use different language. But these economists from Chicago saw the threat of corruption of large corporations wielding public power. They were worried about a future of “rent-seeking,” as they called it, shifting public policy as a strategy for increasing profits. In “The Theory of Economic Regulation,” George Stigler famously wrote that “regulation is acquired by the industry and is designed and operated primarily for its benefit.”34
Stigler, Gary Becker, Richard Posner, and others argued that the size of government should shrink to prevent corruption, because a smaller government with weaker central governing powers would create less incentive for private actors to seek public power. They argued that rent-seeking would be more likely in highly regulated industries because the existence of regulation and differentiation is what inspired corporate political involvement.
However, they did not push for an aggressive antimonopoly strategy. Instead, they pushed to dismantle antimonopoly laws. Why? They made two basic theoretical mistakes in their description of politics. First, they imagined a limited set of policies that might affect a company and, second, they presumed an upper limit of the value that companies could extract from governments. Judge Posner argued that once a company becomes a monopolist, it has “less incentive to expend resources on obtaining the aid of government in fending off competitors” than one in a highly competitive industry.35 Posner imagines that the would-be monopolist faces a single rent (monopoly) that, once secured, sates his interest, and operates as a ceiling of all possible rents. This is clearly false: experience shows that big companies, having invested in securing a foothold in power, will have already paid much of the fixed cost of building the machinery to exercise public power, and will be more imaginative (and efficient) in using it to secure more benefits of different kinds. This logical flaw also shows up in the work of Gary Becker. In his classic 1983 paper modeling rent-seeking, Becker describes an upper limit on what a company will seek from the government: “The total amount raised from taxes, including hidden taxes like inflation, equals the total amount available for subsidies, including hidden subsidies like restrictions on entry into an industry.”36 However, the creative rent-seeker, like the entrepreneur in any area, will not look at present flows to determine potential flows, but will look at possible flows given political limitations. There is no theoretical constraint on the potential size of the subsidy. The potential value of the subsidy is not defined by existing taxes. More taxes can be levied: the existing population of the country does not define it, because levies (direct and indirect) can be brought to bear on other countries’ populations. As a theoretical matter, then, the upper limit of a subsidy from a government is the maximum revenue the government can generate through its power. (As a practical matter, the probabilities approach zero as the subsidy approaches the maximum revenue.) This is not a small point. There are plenty of real-world examples in which companies exercise public power to secure benefits despite the absence of existing revenue. The bailout of the financial institutions is one example; the insurance mandate sought by insurance companies is another. And at a smaller scale, laws that require schools to teach technology classes are, from the perspectives of certain technology companies, rent-seeking laws: they are not grounded in existing revenue but rely on school boards to create it. Deficit spending is not limited by current tax revenues. And one can seek rents through the manipulation of monetary policy in a way that is not limited by existing revenues. In other words, the total potential benefits are bounded by the total potential (not actual) governmental revenue, including debts. The fixed upper limit model was essential to the argument that that concentration in industries posed no corruption threat.
Anticorruption crusaders have for decades asked companies to join them in fighting corruption on a global level. Some of these efforts have doubtlessly produced public good. However, corporate social responsibility is bound to be insufficient to address the threat of corruption that flows from those companies themselves. Even the most aggressive corporate social responsibility standards do not exhort companies unilaterally to become less politically powerful. Even if they did, it is unlikely that such an exhortation would work: it is hard to imagine Microsoft choosing not to merge with LinkedIn because of internal csr policies.
Corporate public corruption flows from a tragic tension: between directors’ or officers’ obligation to the corporation's health, and their ability to increase profitability by increasing corporate power. There is ample evidence that massive corporations, even those perceived as leaders in csr, invest heavily in public relations to reduce their tax burden. They do not bribe, but they extract wealth from the public through tax cuts; on a net level, they add more corruption than they reduce.
One approach locates the institutional flaw in corporate law and corporate obligations, arguing that officers and directors should be ethically free to pursue the public good even when it directly conflicts with corporate goals. In the Aristotelian framework, one might call this the aristocratic approach: the goal is to free corporate ceos to be aristocrats instead of oligarchs. While I laud these efforts, I am troubled by the vision they present: unaccountable corporate actors independently choosing that which is best for the country, and quite possibly the world. Moreover, systems of aristocracy are notoriously weak, and tend toward corruption themselves. Freedom plus exhortation does not always mean virtue. The occasional multinational will resist the temptation to reduce its own taxes or deregulate its industry, but that is hardly a prospect to rely on. As Madison famously wrote in Federalist Paper No. 51: “If angels were to govern men, neither external nor internal controls on government would be necessary.”
The problem is not with the existence of the corporation, or with corporate law. More free and open markets would lead to less corruption. The problem is with concentrated power: a handful of actors who are sui generis; so large and powerful they can bend public power. The modern anticorruption movement chooses not to address these large actors, using formalism or legalism as an excuse, at all of our peril.
As when Siemens spent millions of dollars paying executives to win contracts at a state-owned utility in Israel. “Siemens to Pay to Settle Bribery Case,” Reuters, May 26, 2016, http://www.reuters.com/article/israel-siemens/siemens-to-pay-israel-43-mln-to-settle-bribery-case-israel-idUSL5N17Z2N6.
For example, see Michael Johnston, Syndromes of Corruption: Wealth, Power, and Democracy (Cambridge: Cambridge University Press, 2005), 60–88.
For example, see Amitai Etzioni, Capital Corruption: The New Attack on American Democracy (New Brunswick, N.J.: Transaction Books, 1988).
Aristotle Nicomachean Ethics 8.1.
For some discussion of the debate, see Michael Johnston, “Public Officials, Private Interests, and Sustainable Democracy: When Politics and Corruption Meet,” in Corruption and the Global Economy, ed. K. A. Elliot (Washington, D.C.: Institute for International Economics, 1997), 61–82; Dennis F. Thompson, Ethics in Congress: From Individual to Institutional Corruption (Washington, D.C.: Brookings Institution Press, 1995), 7 [corruption ordinarily involves the “improper use of public office for private purposes”]; Michael Johnston, “The Political Consequences of Corruption: A Reassessment,” Comparative Politics 18 (4) (1986): 459, 460 [corruption is “the abuse of public roles and resources for private benefit”]; and Joseph J. Senturia, “Corruption, Political,” in Encyclopaedia of the Social Sciences, ed. Erwin R. A. Seligman and Alvin Johnson (New York: Macmillan, 1931), 448 [defining corruption as “the misuse of public power for one's own personal profit”]. For an argument that morality is essential to understanding corruption, see Laura S. Underkuffler, Captured by Evil: The Idea of Corruption in Law (New Haven, Conn.: Yale University Press, 2013).
Richard Mulgan, “Aristotle on Legality and Corruption,” in Corruption: Expanding the Focus, ed. Manuhuia Barcham, Barry Hindess, and Peter Larmour (Canberra, Australia: anu Press, 2012), 26.
The idea that private actors might be part of a corrupt system has received attention. Michael Johnston, for instance, has argued that in societies with close ties between political elites and companies, the lines between public and private are blurred, which implicates corruption because corruption depends upon the idea of clearer lines between the spheres. I see this essay as supporting Johnston's arguments about influence markets and elite cartels.
Sung Hui Kim, “Insider Trading as Private Corruption,” UCLA Law Review 61 (2014): 928–952.
To be clear, the problem of private actors corrupting public actors via legal means is central to much of the literature, but I make a slightly different point.
See, for example, Melvin Urofsky's description of Louis Brandeis's battle with Boston Elevated. Melvin I. Urofsky, Louis D. Brandeis: A Life (New York: Schocken, 2012), 135–139.
Trist v. Child, 88 U.S. 21 Wall. 441 441 (1874). Emphasis added.
Much of the discussion about private entities–including big multinational companies like Monsanto, Google, or Siemens–concerns whether they are engaged in what is sometimes called private corruption, and whether we should even recognize the category of private corruption. For instance, on a smaller scale, if a bank manager accepts a personal payment from a customer in exchange for giving that customer special banking treatment, is that corruption? This essay does not address that question, but rather argues that there are many situations in which private actors do exercise governing power, and should be recognized for such. There is also a split between analysts who believe that there is a separate category of private-to-private corruption, and those who believe there is not. For purposes of this essay, I do not engage that argument: I am not addressing those situations in which Siemens might accept bribes from Bayer, but rather when Bayer uses its power to influence the German government.
For a recent volley in this debate, see Kim, “Insider Trading as Private Corruption.”
Joseph Nye, “Corruption and Political Development: A Cost-Benefit Analysis,” The American Political Science Review 61 (2) (1967): 417.
Arnold Heidenheimer and Michael Johnston argue that corruption has been defined in three ways: “public office centered,” “market centered,” and “public interest centered.” See Arnold J. Heidenheimer and Michael Johnston, eds., Political Corruption: Concepts and Contexts (Abingdon, United Kingdom: Routledge, 2000), 3–6. A closer examination, however, shows that the market-centered approach also depends upon formal office office-holding, since it describes corruption as those instances in which entities seek rents from “the bureaucracy” or public officials. For a discussion of the market-centered definitions of Jacob van Klaveren and Nathaniel Leff, see ibid., 8. Moreover, some of the public interest–centered definitions also employ public office office-holding and formal roles. For instance, Carl Friedrich attaches corruption to power-holders tasked with certain duties. Friedrich argues corruption exists when a “power-holder” “is by monetary or other rewards not legally provided for, induced to take actions which favour whoever provides the rewards and thereby does damage to the public and its interests.” Carl J. Friedrich, “Political Pathology,” The Political Quarterly 37 (1) (1966): 70, 74. For a more general discussion of these different approaches, see Heidenheimer and Johnston, eds., Political Corruption: Concepts and Contexts, 6.
The functional approach could consider this business owner not corrupt if he were using his wealth to shape public power to pursue public ends, even if we disagreed with those ends.
The United States’ federal criminal law of bribery has, in some very limited circumstances, taken a functionalist approach to determining who is a “public official.” In Dixson vs. United States (1984), an administrator at a private nonprofit corporation, administering federal grants, took money from a contractor in exchange for steering federal funds the contractor's way. The administrator argued that he should not be convicted under federal bribery law because he was not a public official. The Supreme Court rejected that argument, and instead used something like a functional analysis, suggesting that someone is a public official depending on a fact-specific analysis of the degree to which he exercised public power. However, Dixson involved statutory interpretation, not interpretation of public terms of approbation, and the statute at issue defined a public official as one “acting for or on behalf of the United States.” Dixson v. United States, 465 U.S. 482 (1984).
Stephen Bell, Economic Governance and Institutional Dynamics (Oxford: Oxford University Press, 2002).
The approach that would make the least sense is one that identifies alcoholism by looking at who is breaking laws related to alcohol, using formal means to determine a very difficult and perhaps impossible diagnostic question.
Natasha Singer, “How Google Took Over the Classroom,” The New York Times, May 17, 2017, https://www.nytimes.com/2017/05/13/technology/google-education-chromebooks-schools.html?mcubz=0.
See Dennis F. Thompson, “Two Concepts of Corruption: Making Campaigns Safe for Democracy,” George Washington Law Review 73 (5–6) (2005): 1046.
McCormick v. United States, 500 U.S. 257 (1991). I have oversimplified a complicated area in my description above, but the basic point remains: the requirements for one are different than for the other.
McDonnell v. United States, 579 U.S. ___ (2016).
Zephyr Teachout, “The Forgotten Law of Lobbying,” Election Law Journal 13 (1) (2014): 4.
For example, see Timothy Besley and John McLaren, “Taxes and Bribery: The Role of Wage Incentives,” The Economic Journal 103 (416) (1993): 119.
See, for example, the Transparency International website: “Corruption generally comprises illegal activities, which are deliberately hidden and only come to light through scandals, investigations or prosecutions.” Transparency International, “Corruption Perceptions Index 2014: In Detail,” https://www.transparency.org/cpi2014/in_detail#myAnchor1.
See Susan Rose-Ackerman, Corruption and Government: Causes, Consequences, and Reform (Cambridge: Cambridge University Press, 1999).
It is a variation of the Hobbesian idea that the sovereign controls law and therefore meaning: “The judgement of what is reasonable, and what is to be abolished, belongeth to him that maketh the law, which is the sovereign assembly or monarch.”
See, for instance, Susan Rose-Ackerman, “International Actors and the Promises and Pitfalls of Anti-Corruption Reform,” University of Pennsylvania Journal of International Law 34 (3) (2013): 484. Rose-Ackerman writes, “Although outright corruption in the form of bribes and kickbacks will remain a problem facing all polities for the foreseeable future, those interested in promoting economic growth, poverty alleviation, governance reform, and market efficiency also need to consider how the legal exercise of financial power undermines these values.”
They also have a separate, personal desire to use public power for private ends, since corporate pay is regularly tied to shareholder value; this creates separate issues that I do not address here, but note: it creates conditions of excessive temptation, more than most people can bear.
Burwell v. Hobby Lobby Stores, Inc., 573 U.S. ___ (2014).
For example, see Lynn Stout, “Corporations Don't Have to Maximize Profits,” The New York Times, April 15, 2015.
George Stigler, “The Theory of Economic Regulation,” The Bell Journal of Economics and Management Science 2 (1) (1971): 3.
Richard Posner, Antitrust Law, 2nd ed. (Chicago: The University of Chicago Press, 2001), 23.
Gary S. Becker, “A Theory of Competition Among Pressure Groups for Political Influence,” The Quarterly Journal of Economics 98 (3) (1983): 372.