1 Introduction

Governments are expected to respond to collective risks and mitigate negative externalities emerging from people’s interactions in markets and society, including complex risks such as pandemics and environmental crises. Yet government decisions and rules tend to benefit some more than others, and they impose costs on us. Are these costs externalities? Tullock acknowledged that ‘[e[xternalities are the reason that we must have governments and in many cases they lead to discussion of specific government activities, which generate further externalities and hence require further governmental activity’ (Tullock, 2005a, p.1). Public choice scholars have stressed that the existence of externalities in markets does not automatically justify government intervention because governments can fail in handling them due to cognitive errors or opportunistic behavior (Buchanan & Tullock, 1962; Stigler, 1971; Wolf, 1979; Becker, 1983; Pennington, 2011). In addition, normative public choice presents an exchange paradigm and a framework of consent based on the unanimity principle, which serves as an idealized model against which to judge the distributional outcomes of policies (Meadowcroft, 2014).

In this paper, I provide a definition of government externalities that is useful for both normative theory and a discussion grounded in the actual conditions of contemporary democracies and markets. This definition helps with addressing the question of what societies can expect from governments in view of perceived market failures, collective risks and crises. Defining government externalities is also important for conceptualizing and anticipating government failure, which may extend to the working of markets, since market activities are conditioned by the rules and the rights established, reformed and enforced by governments, including property rights, licenses, taxation and regulations, which blur the conceptual boundaries between market performance and political performance.

The paper starts by presenting the conceptual problem with the definition of externalities as follows. If we see ourselves as complete outsiders of the political decision-making system, most of what the government does inevitably creates ‘externalities’ simply because some of us will hold minority views about these government actions and some of us will pay more than what we benefit from. However, by the same token, it could be said that all market operations reproduce these ‘externalities’ because market operations rely upon on a set of rules and institutions produced by contested processes of government rulemaking. It is public institutions that define property rights, enact them, and occasionally revise and reform them, and some of us may not like the rights we have against what others enjoy. Because property rights determine who can act and transact in markets, as well as who can bargain around the costs one incurs from the activities of others, each arrangement of property rights can be politically contested, and almost everything can be subjectively perceived and publicly condemned as an ‘externality’ in this broader sense.

The diametrically opposite position would be to see ourselves as participants in government decision-making: democratic citizens with civil rights to contest and revise government rules. In that view, nothing that a democratic system of laws generates is external to our decision-making position. The third avenue is to follow what constitutional political economy in the tradition of public choice posited (Buchanan & Tullock 1962). If the government acts in conformity with the rules of a constitution agreed upon by unanimity, then government actions can be seen as internal and not external to the scope of unanimity, and therefore this is a way of claiming that government decisions are not creating externalities insofar as they stem from within a broader framework of mutual agreement.

The paper discusses government externalities with reference to both these positions: the non-ideal circumstances of democratic participation and Buchanan and Tullock’s normative constitutional theory. No matter what normative position we take, if we look at how politics operate, the answer to the question ‘Is there such thing as government externalities’ is a clear ‘yes’ because many, if not most, consequential political decisions are primarily shaped by agreements outside the formal structure of public rights citizens enjoy and can exercise. These agreements take place behind closed doors in the form of clientelist transactions for the distribution of private gains from government actions to a selected few, and they generate costs for the rest of us which are not authorized by the system of rules that makes us participants in the decision-making system. I show that governments (1) create a type of dark externalities – stemming from opaque clientelist deals whose cost is borne by the public, and (2) this type of externality affects how markets operate and how governments handle other complex risk situations under incentives embedded in politics that allow some privileged actors to exploit public decisions for the purpose of achieving private gain.

Clientelist exchanges are vital for the success of politicians and political organizations. Politicians tend to give priority to the demands of influential socioeconomic actors and groups in return for political loyalty, firm support and sizeable resources. This creates a structure of informal exchanges that permeates and erodes the formal structure of decision-making rules and stealthily shapes decisions over the provision of public goods and services. To the extent that some actors manage to capture and trade political power, any system of governance, even a democratic system, will generate costs that occur contra the constitutional provisions that create legitimate expectations of how government decisions are supposed to be shaped by our participation. These costs are government externalities.

Government externalities are also complex because they create higher-order costs beyond the sum of the fiscal cost of the individual transaction. The added cost involves regulatory distortions, more red tape, barriers to market entry and burdens on the operation of markets that distort the provision of public goods and services. In that sense, government externalities are a case of government failure by analogy to how some market externalities can be considered a case of market failure. They also affect the way governments are addressing social risks, such as pandemics, economic crises, banking crises, and environmental problems. Moreover, these distortions can be manifested as perceived cases of market failure, in the form of a significant mismatch between the supply of goods and services and shifting demand, which market actors cannot resolve and correct. Broadly speaking, clientelism can be considered a case of institutional failure because it affects both policy performance and market performance simultaneously and in combination. Finally, government externalities are an inherent byproduct of the political process and, as such, they can hardly be eradicated through political action and institutional reform.

2 The concept of externalities and the blurred boundaries between markets and government

A simple definition of a negative externality can be taken from any undergraduate textbook. To start with, an externality is defined as a cost or a benefit to a third party of an exchange or activity in which they do not participate. However, defining participation is problematic with regard to who decides and who has actual control over that decision. Buchanan and Stubblebine (1962) give a general definition of ‘an external effect’ as a situation when the utility of an individual, A, is dependent not only upon the activities that are exclusively under one’s own control or authority, but also upon another single activity which is under the control of a second individual, B (ibid. p. 372). Yet what they mean by control is not specified. Should we consider the costs or benefits small shareholders receive from executive decisions in a firm as an external effect for them? If the term externality applies to those who have no consequential control over a decision, then the answer will plausibly be ‘yes’. But, in reality, shareholders have decision-making rights, which they can exercise inside the firm, even if some of them sometimes do not succeed in persuading others, and do not ultimately control decisions that affect their circumstances.

In Meade’s definition (1973) an externality is more narrowly defined as an event that confers an appreciable benefit or inflicts an appreciable damage on some person or people who were not fully consenting parties in reaching the decisions that led directly or indirectly to the event. Lack of consent makes them non-parties in decisions made by those who exercise a right to transact or act. However, one’s position regarding consent granted for a decision is determined with reference to where one stands in terms of the decision-making rights that apply to that decision. In markets, property rights allow contracting parties to agree on a transaction. Third parties are those who do not have property-derived decision-making rights. A system of laws establishes what is recognized as ‘mine’ and ‘yours’ in exchanges of goods and services, and who has the right to decide on these exchanges while all others remain third parties with rights potentially affected by that transaction. Based on one’s position in this system of rights, we can credibly say to others, ‘this is my property to sell’, ‘my money to use’, as well as ‘this is my property whose value has been diminished as the result of your decision’. After all, only ascribed rights allow effect – as in damage or cost – to be recognized, consent to be given for their use or exchange of goods and services, and, value, benefit or cost, to be calculated.

Indeed, the best approach to the definitional problem of externalities is to pin externality to a system of rights which, in markets, allows for transactions to take place and value to be calculated (Cornes & Sandler, 1996, pp. 40–42). Coasian bargains over externalities are then possible (Coase, 1960) but their outcome depends on an original distribution of property rights among the involved actors. If high transaction costs prevent many actors from privately negotiating a solution over a type of an external cost and reaching an agreement, in that case, tackling transaction costs and allocating property rights can become a source of public dissatisfaction and, often, the ground for political action calling for the government to change the rules and the content of property rights. Moreover, because markets are embedded in rights and these rights are embedded in institutions, some may also question the outcomes of market operations more generally, as well as challenge the very specification of property rights given their consequences. Here, we are entering the turf of politics and the processes of institutional design and government rulemaking.

What is more, Buchanan and Stubblebine (1962) note that not all perceived externalities are worthy of institutional action and that the externality becomes ‘potentially relevant’ ‘when the activity, to the extent that it is actually performed, generates any desire on the part of the externally benefited (damaged) party (A) to modify the behavior of the party empowered to take action (B) through trade, persuasion, compromise, agreement, convention, collective action, etc. (Ibid pp. 373–374.) Buchanan and Stubblebine (1962) add that the desire to modify the behavior of another does not imply the ability to implement this desire. They define an externality as Pareto-relevant when ‘the extent of the activity may be modified in such a way that the externally affected party, A, can be made better off without the acting party, B, being made worse off’, which is not always the case (Buchanan & Stubblebine, 1962, p. 374). This is a response to Coase pointing out that there is the possibility and a scope for negotiations between the two interested parties when compensation can be paid by the affected party to the party who is acting within the scope of their property rights. What Buchanan and Stubblebine say is that not all externalities can be fully negotiated away, and tackled by the private parties themselves.

Reading Coase and Buchanan and Stubblebine together suggests that economic actors will inevitably have residual preferences that involve political action and institutional redesign concerning transaction costs and the original distribution of rights, because this context matters as to who is entitled to act and who can ask compensation, and because, even if they are the ones to receive compensation through negotiation, they may still incur a residually negative external effect.

Because the allocation of private rights determines prospective costs and benefits among those who wish to act and those who want to protect themselves from the actions of others, even the original allocation of rights inevitably become the subject of social contestation and enters the realm of politics, for instance, ‘how much can I build in my plot of land?’, ‘where can I build a factory?”, ‘how tall my building will be?’, etc. Governments are often called to settle these disputes by enacting or reforming rights by way of legislation and institutional reform.

When fixing external costs becomes the scope of government action and the turf for political competition, even pecuniary externalities matter (Cf. Viner 1953), which emerge when private actions cause an increase or decrease in market prices, potentially reducing someone else’s profit. This is because profit losses can motivate some actors to appeal to government for regulatory interventions that will protect their businesses from exposure to competition (as in the case of hotels versus Airbnb, for example) by changing the terms in which products and services are provided. It is important to note that pecuniary externalities from market operations, are not typically classified as externalities – in fact there are an indication that the market is working competitively, and the very term has been criticized as redundant (Chipman 1965, p. 746). By contrast, negative externalities that are regarded a case of market failure are cases in which actions influence the consumption and production possibilities of others in ways that the price of the product fails to incorporate those costs, often reducing the ability of others to produce or consume goods and services in quantities that would have matched demand, for instance, by causing pollution or poor health. However, this conception is derived from a type of economic theory that anticipates a social optimum market output based on which theory can identify hypothesized mismatches between marginal private costs and marginal social costs and could characterize undersupply or oversupply as market failure. In reality, all external effects are contested and become the basis for demands for government regulations and redistribution.

In that light, the very notion of market failure can be problematized, given that markets and politics are entangled at the level of institutional rules and political interventions. In the absence of the ideal conditions of neoclassical analysis, a good or a service is undersupplied against a constantly shifting level of demand within a context of rules and institutions affecting production, provision and consumption of goods and services. This undersupply may be mitigated or removed thanks to innovations stemming from within a competitive market process, when entrepreneurs manage to create a private (excludable) good and set up a market for it. When there is a consistent poor record of markets matching demand, most probably because of the nature of the good as non-excludable (Musgrave, 1959), and no technological, institutional or entrepreneurial solution has been devised to change this situation, this can then be called a market failure eliciting an institutional solution (Ostrom, 1990). In this context, system-level failure can be defined as the degree of deviation between the actual performance of the market and the expectations about how a system of property rights would reasonably operate so that supply better tracks and satisfies an evolving demand for a good or a service.

It thus becomes obvious that the boundaries between markets and governance are porous. The solution to a perceived problem of undersupply, for example, may be an adjustment of institutions, which includes private negotiations, the possibility of government interference as well as other institutional innovations (Cf. Coase 1960; Ostrom, 2010; Paniagua & Rayamajhee, 2022). Expanding on the work of Elinor and Vincent Ostrom (2002, 2010), Rayamajhee and Paniagua argue that the ‘inherent’ nature of goods and their specific taxonomy are not static and definitive concepts but are instead contestable and dynamic features that are institutionally contingent’ (Rayamajhee and Paniagua 2021, p.71). ‘[A] typical private good, devoid of the institutional context characterized by private property rights, rule of law, and low exchange costs, is no longer excludable and subtractable (i.e., private)’ (Rayamajhee and Paniagua 2021, p. 72). Cowen (1985) argues that the publicness (or privateness) of a good is not a characteristic of each good itself but is defined by the attributes it has been given by the institutional context.

Hence, it is problematic to reduce the social order to a binary distinction between markets and governments, given that negative externalities and the concept of market failure can be seen as a question of institutional design, policymaking and market innovation. Both the concept of failure and the notion of externalities are not independent from the circumstances of politics. Broadly speaking, market failure and externality must be seen as problems subject to institutional redesign, political contestation and agential innovation. Politics is present in all proposals and efforts to correct suspected market failures and handle externalities. Keech and Munger (2015) go as far as to argue, because governments enact and enforce structures of property rights, government failure is ‘causally prior’ to the operation of markets, and markets fail because the government has failed to better define the rules under which markets operate (ibid., p. 3). Furton and Martin (2019) suggest that this situation can better be described as ‘institutional mismatch’, when the rules governing an economic problem are inferior to a feasible alternative set of rules.

3 Defining government externalities

Because the process of institutional reform is politically and ideologically contestable (Rayamajhee, 2020), idealized conditions of perfect governments serve no purpose when it comes defining government externalities either. As Demsetz (1995) stresses, ‘The mythical state must not be compared to actual markets in which negotiations and information are costly; nor should the mythical perfect market be compared to actual political institutions (ibid. p. 578). Like markets, there are no ‘perfect’ conditions of governance to plausibly sustain a definition of government externality against a demanding and unrealistic benchmark that public goods can be efficiently provided in full alignment with public demands. This can never happen in politics due to diverse preferences, interests, and ideas as well as the informational overload from ongoing preference expressions, inadequate and knowledge over cause-and effect regarding government actions, and the presence of several trade-offs concerning the use and allocation of scarce resources.

So, here’s the major puzzle. If almost everything the government does generates an asymmetrical and contested distribution of costs and benefits, there can be no Pareto efficient outcome from hardly any government action. Can we then consider all government actions as creating ‘externalities’ by analogy to the concept of market externalities? The answer depends on which criterion we choose according to which citizens can be considered as ‘consenting’ participants in government decisions and not third parties.

In democracies, government actions are supposedly legitimated by inclusive decision-making processes. Consider an example. Welfare-spending asymmetrically benefits some and burdens others, and involves government decisions authorized through the exercise of our democratic rights. While citizens may have diverse opinions of what level of spending is good for them personally or in general, the constitutional system has determined the procedures by which these views can be expressed, ideally creating inclusive and transparent processes of decision-making and policy selection in view of legitimizing policymaking through public participation. Democratic legitimacy is the public recognition that democratic systems establish a system of public rights that gives citizens civil rights for having some input into political decision-making. These rights allow each citizen to register their approval or disapproval periodically through elections, and through other forms of expression and collective action. Democracy is essentially a system that both recognizes and seeks to handle preference heterogeneity, which is also present in markets, but acknowledges that, unlike markets where diverse preferences can be satisfied by diverse suppliers, government decisions are all-encompassing and binding for all, erga omnes, for social environments characterized by diversity and conflicts of preferences and ideas. With voting rights and freedoms of expression, petition and protest, as well as with the right to challenge the legality and constitutionality of government decisions, democratic systems seem to position every citizen as an insider to the political decision-making process. Constitutional rights also affirm a protected sphere of private rights, which allow us to make lawful claims of protection against unauthorized state actions, with court decisions, rulings and case-law interpretations acting as checks on government authority.

Challenging this democratic-majoritarian view of participation, normative public choice adheres to a voluntary exchange tradition and examines unanimity as the standard for original consent in a system of collective decisions. If we believe that subjective evaluations determine what ‘socially ‘efficient’ allocations in politics mean, this establishes a strict criterion: only unanimity eradicates externalities. Assuming that an agent will vote for a government proposal if their overall costs are outweighed by the benefits (Buchanan 1968: 92), unanimity achieves Pareto efficiency. According to this benchmark, if no consensus can be reached regarding a change from an initial position in terms of rules and allocations, any deviation from that position will result in a decline in utility for at least one person, which is Pareto inefficient from the perspective of subjective preferences. What is quite interesting is that, as a steady state, we can consider the initial position as Pareto efficient (Buchanan 1967: 285). However, this does not mean that there has been earlier unanimous consent over the enactment of that policy at some time before and, therefore, in a way analogous to market externalities, if we adopt a dynamic view of how government policy is continuously generated, we cannot do away with the question of externalities.

Buchanan and Tullock (1962), recognized that no institutional system of governance can be set up to work with unanimity at all times and, adhering to normative individualism as the foundation for a liberal project in constitutional political economy, proposed a normative constitutional theory on the basis of a distinction between constitutional agreement and non-unanimous post-constitutional decision-making. At the stage of constitution-building, citizens are presented as rational political actors who seek to determine, in a contractual way, which activities can be run by government or will be regulated by it based on a calculus in which the relevant benefits and costs of ‘collectivization’ can be compared versus the relevant costs and benefits of allowing private voluntary action. This is an idealized view of a constitutional process endogenously creating the system of rights and procedures that will later determine the legitimate type and range of costs and benefits which its operation will generate. The constitutional framework agreed upon by unanimity will provide the socially – intersubjectively speaking – optimal rules for collective decision-making operating and allocating costs and benefits. These allocations will be covered by the constitutional consent. This is a way of considering externalities as eradicated from the system of government from this intersubjective perspective as described above. Once there is agreement on property rights and the rules of the post-constitutional stage, these decisions are not creating externalities insofar as they take place in accordance with a broader framework of mutual agreement.

To sum up, whether the costs of government will be seen as externalities depends on which model of consent we use as an evaluative framework for eradicating externalities. There are two distinct models to serve as criteria: the strict unanimity-based criterion of public choice and the democratic participation model, which is rather imperfect but practically focusing on democratic participation as if it implies and conveys some form of tacit consent for government decision-making. In other words, for either models, the term externality would only plausibly apply to costs imposed on us outside the processes each model understands as sufficiently inclusive participatory and covered by public consent. Otherwise, even if we may find ourselves on the side of a dissenting minority for a public decision (Cf. Tullock 1998, p. 412), our position as dissenting citizens can be paralleled to shareholders who have a minority position against a decision approved by those who hold the majority of shares, and against decisions of the executive board that affect them negatively and even cost them money. Unlike citizens in democracies, shareholders have explicitly signed up to this structure of rights and have the freedom to exit the firm. Unlike citizens, however, their voting rights are not equal but depend on the number of shares they own. It suffices to say that being just a weak participant in the decision-making process, for example, a mere voter, is not synonymous with the status of being a complete outsider.Footnote 1

4 Clientelism as a source of government externalities

It is worth exploring which type of political actions and transactions falls clearly outside a decision-making framework that involves participation and can invoke consent, whether we believe that this is an inclusive democratic process or whether we subscribe to Buchanan and Tullock’s normative approach in which unanimity at the constitutional level can tackle this problem by creating a post-constitutional order covered by consent. In either cases, externalities will still emerge from patron-client exchanges that shape policy outcomes against how, in each perspective, a legitimated system of public decisions is supposed to work.

Clientelism describes the distribution of resources and benefits by political actors – the patrons - to their clients through an agreement hidden from the public in which politicians make this allocation dependent on the political support of the beneficiaries (Trantidis, 2016, p. 6). Politicians and policymakers are most likely interested in serving their own agenda first, whatever that may be, for example, getting re-elected, building a political career, securing a private career after politics, helping their friends and family build a career, or simply becoming richer. Politicians may explicitly ask their clients to make contributions to their campaign in the form of money, activism, media support etc., in exchange for access to government-distributed goods. Socioeconomic actors may then offer campaign resources explicitly just like bribes paid to those who hold government power to achieve favorable government actions and policies, such as licenses, insider information, regulatory concessions, protectionist measures, subsidies and favorable law.

Clientelism is a special instance of particularistic politics. In democracies, utility considerations are shaped in two key competitive arenas, on the one hand, economic actors and social groups competing for benefits distributed by the government and, on the other hand, political actors competing for election to office. The democratic constitutional order does not elicit that policymaking must only be serving ‘a public interest’, and does not equate political access with substantive political equality. Institutions allow the generation of dispersed costs and concentrated benefits legitimately because the understanding is that people have diverse preferences and conceptions of what the government should do. Government interventions, distributions and regulations are largely driven by private interests anyway (Olson, 1965; Tullock, 1967; Peltzman, 1976; McCormick & Tollison, 1981). Access to political power is also unequal in real terms. Small and organized groups tend to more easily organize action to claim benefits from politics (Olson, 1965) sometimes simply by providing information that changes how policymakers view an issue and try to solve a problem (Potters & van Winden, 1992). Unequal political access here still conforms to the system of public rights in place. For example, strikes, protests and rallies are lawful exercises of public rights by those seeking to exert pressure on policymakers.

However, clientelism is a particular form of exchange that violates the public system of decision-making when the office holders agree to satisfy special interests in return for benefits at the dark backdrop of formal processes. This distinction clarifies why this practice systemically generates externalities. Some actors or groups choose to trade resources with politicians and political organizations in exchange for state-provided benefits behind closed doors and outside formal and transparent political avenues (Cf. Stigler 1971; cf. Bartels, 2008; Bonica, 2013; Rigby and Wright 2013; Barber, Canes-Wrone, and Thrower 2017). The way these deals are reached breach the rules of the system that allows citizens to be observants and, to some extent, participants in political decision-making. While discerning citizens can suspect that clientelism has actually shaped several government decisions and actions, the exact terms and conditions for each decision are hidden. In that sense, clientelism is ‘hiding in plain sight’, for example, when lobbying organizations offer campaign donations but, behind closed doors, agree on privileged treatment by politicians in the legislative and the executive branches of government (Cf. Denzau & Munger 1986; Grossman & Helpman, 1994). These exchanges are not taken in the visible realm of democratic norms and in conformity with rules and procedures for public participation. They produce costs which, probably, most of the public would have disapproved of had they have the chance to be informed about these exchanges and have a say on them.

In short, clientelist exchanges violate basic expectations of how the formal democratic rules and processes secure citizens participation, inclusion, and public accountability. Mutual gains accrue to the parties of this type of transaction (a positive-sum game for clients and patrons), but the government decisions serve purposes other than the ones publicly projected under the formal processes, and impose costs on others, partly or wholly, on false pretenses. They thus meet the criterion of a negative externality as they shape public policy and result in public spending, redistribution and regulations whose cost we then incur as third parties.

Looking at clientelism as a source of government externalities can be juxtaposed with the term political rent seeking. Political rent seeking is typically presented as distortion of the ideal market conditions for exchanges creating competitive market prices. For Tullock (2005b) rent-seeking is a concept applicable to cases in which individuals and organizations expend resources on lobbying government for special privileges that reduce the wealth of society and have negative social impact’ (ibid. pp. 9; 51). The problem with this approach is that we never have a perfectly competitive market, and, in addition, all transactions rely on allocations of rights by institutions which may already be contested as privileging some at the expense of others. Rents can stem from the way governments have fixed the institutions themselves governing the terms of several market transactions. What is more, in an intersubjective definition of welfare, we must acknowledge that citizens have different and often conflicting preferences and diverse opinions and ideas over what welfare means to them and to society as a whole. While, for Tullock, rent seeking is ‘a negative sum game, as individuals and groups invests resources in attempts to obtain transfers or to resist transfers away from themselves’ (Tullock, 2005b p. 31), we can recognize that every institutional setting entails and allows transfers of resources in different ways.

A focus on clientelism broadens our conception of how much the satisfaction of some preferences via government and the subsequent generation of costs and benefits can be out of line with the formal decision-making rules and processes according to any criterion of public participation and institutional consent we may subscribe to. Practically speaking, government externalities are those costs from government decisions not authorized by the public rules that set up relatively inclusive and accountable decision-making processes. In the next section, I will explain why government externalities are complex and how they are related to government failure and market failure, positioning both these notions in the context of an institutional analysis that acknowledges the entanglement between markets and governance.

5 Government externalities as complex externalities

The concept of government externality is also helpful for curbing our expectations regarding the capacity of governments to tackle other externalities, such as risks, perceived market failures and perceived negative externalities from market operations. Complex problems include pandemics, large-scale environmental crises such as climate change, ocean and waterways pollution, plastics and microplastics in our ecosystem, large-scale flooding etc. Paniagua and Rayamajhee (2022) use the term ‘complex externality’ to refer to problems of defining and creating tradable property rights in those situations, due to the large scale and magnitude of the external cost that usually encompasses various organizations and governments at different scales.

It is important to emphasize that a complex externality involves a synergy of multiple and continually occurring activities producing a sizeable outcome whose effect, scale and severity cannot be simply disaggregated into each specific additive contribution. Consider noise and pollution from cars. A typical car passing by will have a negligible effect on those of us who may stand by, but the traffic created by all the cars in an area of a city generates a damaging level of pollution that will affect those who live and work in that area. The overall damage or risk is a complex externality because it is the emergent effect of car traffic. Scale and synergy are what makes the effect of pollution (more) significant. Even if the pollution caused by each car can be measured, this only matters as a contribution to a total level of pollution produced from all other contributing sources in a geographical area. With complex externalities, there is upward causation from actions to effects, but these effects are relevant and troubling because of their emergent properties. There is then downwards causation from these emergent properties on many of us of not all of us. This cost is higher because scale matters for creating the synergy of actions that inflicts costs and damages to our protected rights, such as health, property or safety.

Indeed, as the scale increases, non-tradability becomes an issue concerning the question of how societies and markets can tackle these complex externalities. Paniagua and Rayamajhee (2022) present situations, such as CO2 production and climate change, in which internalizing externalities is extremely difficult if not infeasible, because the costs of internalization are high relative to the individual benefits, there are limits to scientific knowledge and technology required to mitigate the externality at a large scale, and the externality is spanning different jurisdictions (Ostrom, 2012).

Acknowledging the idea of complex externalities as an emergent effect of activities also concerns the question of what governments must preferably do in those cases. For instance, traffic restrictions in city centers where car traffic is high (for instance, a congestion charge in London), rather than a general ‘environmental’ tax on fuel is a more effective way to reduce pollution and its harmful effects. What is tackled here is not the environmental damage each car generates, which is negligible, but the higher ‘emergent’ outcome that congestion creates. With complex externalities, institution solutions concerning policy regulations and market alternative settlements, whether they create grounds for compensation or whether they are trying to internalize an externality, for example, in the form of a Pigovian tax on gasoline, must be justified with reference to how they tackle the synergy of contributing factors and the overall effect of damage.

In a similar vein, knowing that policymaking also generates externalities can inform our views regarding the capacity and limits of government intervention in tackling complex externalities. Public actions serving clientelist agreements can be justified as interventions to supply public goods and services. Clientelist motives can be masked behind the pretext of public goods provision and behind responses to perceived market failures and complex externalities. Because clientelism is intrinsic to policymaking, an inevitable by-product of political competition, clientelist supply will, most often, take precedence over considerations of public goods provisions, stealthily ushering a ‘clientelist bias’ into policymaking supposedly designed under different, ‘public’ justifications competition (Trantidis, 2016; Cf. Kogelmann 2021).

At this point, it is important to stress that popular expectations of what public policy can achieve tend to ignore or downplay the fact that political decisions often hide clientelist deals, and that clientelism is relevant to how markets perform and deliver goods and services. For example, applying the standard assumptions of microeconomics to governments Wittman (2004) shows that political competition for votes push governments to generate efficient outcomes similar to those of markets. Yet Wittman assumes that there is only one competitive process in politics: competition for votes, which leads to political supply for voter’s preferences judged against an unspecified standard of population welfare. Leeson and Thompson (2021), by contrast, gives a long catalogue of cases where interest groups sought to exert influence on government decisions and secured special treatment in the form of subsidies or favorable legislation in one single sector alone: public health. These informal deals come from a ‘parallel’ market for access to government decision-making. Many goods that appear ‘public’ or ‘private’ may be overproduced because of these deals. This means that governments, when they seek to address social risks and perceived market failures, they may contribute to the creation of new problems and issues.

Hence, government externalities are complex externalities by their very nature. The external effect of clientelism as a general practice at any given time is greater than the cost of all the transactions that have benefited the clients. The full cost of clientelism is hard to measure, because it includes not only the sum of the fiscal costs of these transactions but several institutional distortions that emerge from this practice in terms of who has full market access, how we are treated by the government in distributions and regulations, how much and what type of red tape we must face etc. The institutional rules which governments set for markets to operate can be distorted by clientelism that imposes complex government externalities on us in the sense described above. Calomiris and Haber (2014) as well as Paniagua (2021) present cases of political bargains that led to the formation of a dysfunctional and fragile banking system that was more prone to collapse when a liquidity crisis hit. Systemic fragility in the U.S. financial system (and, consequently worldwide) can be seen as a complex externality largely attributed to these political bargains. The prevalence of clientelist exchanges can go as far as to transform an economy into a limited access order (North et al., 2009). In the words of North, Wallis and Weingast (2009, p. 39):

‘The characteristic way in which a social order structures human organizations also produces predictable features of the larger society. Limited access orders exhibit systematic rent-creation, market power, privileges, and differences between elites and others; they also preclude thriving markets and long-term economic development. Open access orders exhibit systematic competition, entry, and mobility; they also foster thriving markets and long-term economic development.’

When policies or even institutions create distortions that have not been created through full public approval via formal participatory processes, but were largely the outcome of stealthy clientelist exchanges lurking behind public justifications and government actions, we can talk of government failure and institutional failure, as the next section explains.

6 Clientelism as government failure

We have so far recognized that clientelism breaches the norms by which the formal institutional system is supposed to be working, creating ‘dark’ negative externalities at multiple scales affecting, through their complex effects, almost all areas of socioeconomic activity. Is this a case of government failure? In a technical definition, government failure can be said to occur when government interventions worsen, rather than improve, market allocations of resources away from an ideal standard of allocative efficiency, primarily with reference to the preferences of consumers (cf. Grand 1991). But as argued earlier, establishing a standard of allocative efficiency as the basis of comparison is problematic because no market transactions occur outside a government-regulated framework. The so-called competitive market process works in an institutional structure already configured by government and politics.

The previous definition of government externalities helps us build a more realistic conception of government failure in a way analogous to market failure. Externalities from private transactions are considered cases of market failure insofar as they add external costs on non-participants which affect their capacity to produce a good or a service in a given competitive setting and thereby create a mismatch between expectations that demand for a good or a service will be largely matched by adequately corresponding supply. In a similar vein, government failure occurs when similar public expectations for the provision of goods and services (by markets and governments) are not met because additional costs and distortions have been created that were not fully legitimated through the formal structure of public participation. This mismatch can affect how markets operate too because government externalities pass costs on market actors directly or indirectly by the way government has configured the institutional setting for market activities.

It is important to remember that government externalities concern both policy interventions and institutions governing markets. Because real markets operate within a framework of government, there is no price mechanism outside this framework that supposedly brings efficiency regarding the costs of producing an activity generating outputs and profits that stands outside of the circumstances of politics (Cf. Wolf 1979, pp. 63–64). If we see politics and the economy as one entangled sphere of activity (Wagner, 2016), it is also difficult to conduct the thought experiment that Munger (2003) asks us to do: defining the outcome of distortions caused by special interests through a comparison of outcomes: what does the world of special interests look like, and how does it differ from a (possibly nonexistent) world in which special-interest groups play no role in politics?

What is worth stressing is how problematic it is to attribute failure to markets alone without considering how the market works against government rules and the circumstances of politics and, subsequently, as activities affected by government externalities at all times. Market supply is inextricably linked to how government regulations and state interventions set up the terms governing production, transaction, sale and consumption.

Nevertheless, realistically speaking, we can only see government failure as stemming from a mismatch of plausible expectations from how the formal institutional setting must be agreed upon so as to configure market transactions, and what the presence of government externalities add in terms of creating a distance from this expectation. This is not the same as to say that governments are generating costs on production and consumption because this is expected and inevitable. As also stressed above, Pareto efficiency in policymaking and institutional design is impossible but deviations from it can be legitimated by reference to a constitutional model of consent. Instead, the notion of government externalities is important because it illustrates a particular way in which government affects the production and provision of both private and public goods and services outside such a model. A market operating under a system of governance dominated by clientelist exchanges adds costs affecting production and consumption that are not legitimated through our participation in established formal processes. Clientelism illicitly satisfies the preferences of a few, mostly small networks of privileged players, a ‘preselectorate’ (Trantidis 2016, p.225; Cf. the ‘selectorate theory’ of Bruce Bueno de Mesquita et al. 2003). This is a case of government failure because, simultaneously, (a) these costs and benefits are not legitimated by any notion of consent attached to an inclusive decision-making process and (b) they negatively affect the production and provision of goods and services in ways similar to how market externalities can do so.

What is more, clientelist exchanges are not transactions whose external costs can be managed by a Coasian-like negotiation between the transacting parties and the rest of society that incurs its costs. There is no ‘market solution’ here because clientelist deals are not just decisions external to us, but they are also part of processes and deals hidden from us to a significant extent, and not recognized or legitimized through the constitutional system. Their overall cost cannot be identified anyway for compensation to be calculated or offered.

The fact that contra any constitutional provisions creating legitimate expectations of public participation, clientelist exchanges will still take place at all levels of political decision-making is a fact that matters for normative theory too. Clientelist exchanges affect the very process by which constitutional agreement is reached which, in normative public choice, was supposed to be set to eliminate externalities from post-constitutional decisions. Buchanan and Tullock (1962) were hopeful for the effect of a constitutional process tackling the ‘Pareto efficiency’ problem of politics in the agreed post-constitutional order. However, post-constitutional decisions will carry on producing government externalities insofar as there will be clientelist exchanges in the backdrop shaping decision-making not covered by consent and the system of rights created by consent for the post-constitutional order. Constitutional political economy, while creating a fitting analogy for the political process from individual exchanges that occurs in the market, loses sight of a large range of subsequent interactions among actors engaging in bilateral or multilateral exchanges to derive special benefits. While public choice stresses the importance of decision-making rules, clientelism as a phenomenon ingrained in the logic of exchange calls attention to practices infiltrating and corroding the operation of formal rules, by which actors can bypass formal rules and constraints to achieve outcomes through informal avenues not legitimated by the formal system of decision-making.

Taking stock of clientelism enriches our understanding of political behavior in both ideal theory and non-ideal theory, cutting across all levels of political interaction and affecting diverse forms of political organization. Unlike the hopes of early public choice theory, it suggests that formal rules are further away from determining behavior and outcomes predictably. Clientelism is a practice parasitical to any decision-making process, operating outside the expectations of how formal rules should condition these processes. In other words, politics and government are a two-tier game: the formal political process determined by the formal rules and a market of informal exchanges market for the few. While for Buchanan and Tullock (1962), the nature and extent of costs generated by political decisions in the post-constitutional order depends on the formal decision-making rule, this paper stresses that the nature of these costs largely depends on the degree to which actual decision-making breaks away from formal decision-making rules that invoke public participation and make a claim of consent. Even if the constitutional rules are set up specifically to ensure transparency and public inclusion, post-constitutional political decisions will inevitably host a ‘dark market’ of clientelist transactions not authorized by the institutional system and imposing costs on everyone else that can be properly considered as externalities. In other words, politics as exchange’ also emerges in opaque bilateral or multilateral exchanges at the exclusion of others and no constitutional design can eradicate informal exchanges from how the post-constitutional order will operate.

What is more, the very process of constitutional design can be ridden by clientelism. Cowen (2018) notes that, quite often, small groups can succeed in influencing the constitutional process and alter the constitutional framework in ways that reduce the opportunity for co-operative production and create more opportunities to reward themselves and their allies at the expense of their competitors and adversaries. Hence, even an ideal constitutional process that conforms to normative contractarian rules may host exchanges among participants whose preferences would involve a desire to embed clientelism in the post-constitutional order. This means that a turf for clientelist exchange can be sneaked into the post-constitutional rules, and this can happen in conditions of imperfect information: some participants may deliberately want it because they anticipate special gains while others will not discern the clientelist preferences behind some ideas for constitutional reform. Indeed, while factionalism and group domination puzzled the founding fathers of the early American Republic, their constitutional design seems to have overlooked the full range of clientelist exchanges that later developed system-level effects changing the nature of the political and economic structure of the United States.

A plausible question is the following. If clientelism is ingrained in political decision-making, is there any institutional solution to decrease its extent? Different institutional arrangements, such as majoritarian democracy, presidential and parliamentary systems, qualified-majority rule etc. introduce different veto players. These arrangements can shift, either adding or removing, checkpoints and avenues for clientelist exchange, and may have a subsequent effect on the ‘prices’ and ‘transaction costs’ of clientelist exchange. Moreover, Trantidis and Cowen (2020) point to the nature of state interventions that are more amenable to this form of exchange by drawing a conceptual distinction between partial-discriminatory and impersonal applications of state interventions. Partial-discriminatory interventions are usually policy measures and commands, such as protectionist measures and subsidies, where it is it relatively easy to identify the winner and use them for the purpose of clientelist exchange.

Unfortunately, the prospect of institutional reform to limit the practice of clientelism confronts the problem of collective action (Trantidis, 2016). Very few would be willing to commit resources, time and energy seeking reforms that will eventually benefit the many. Instead, it is more rational for socioeconomic actors to use these resources to enter a clientelist network, reinforcing rather than undermining the clientelist system. At the same time, from a game-theoretic point of view, clientelism is the dominant strategy among the players of politics: the continuous preservation and expansion of clientelism supply makes sense for patrons and clients, regardless of any external costs on the economy and society. No political force or individual politician involved in this practice would probably gain from withdrawing from this practice or seeking to reform against it, as this will give a competitive advantage to their rivals who want to continue with this practice (Trantidis, 2016). The practice of clientelism can be presented as a Nash equilibrium in politics, a situation in which any unilateral move by a political force in government to reduce the scope and intensity of clientelism is bound to favor the opposition party (Trantidis, 2016).

7 Conclusion

In weighting the costs and benefits from government intervention in complex situations like pandemics and environmental crises, defining and understanding government externalities is important for a discussion about how societies will design institutions and rules governing the operation of markets and best try to mitigate collective risks and handle complex crises.

To that end, this paper points to clientelism as the source of government externalities emerging from all forms and processes of political decision-making. Clientelist exchanges are hardwired in government decision-making and policymaking. Clientelism plagues the design of policy decisions that are taken under a line of public reasoning but hide other true motivations and goals. It also distorts the design of the institutions and norms that govern all other socioeconomic activities. This practice transforms politics and private markets into a terrain of political exchange for special benefits. In that regard, the distributions and regulations shaped by clientelist deals generate externalities on us who find ourselves to be complete outsiders to these clientelist deals. Their overall effect and costs on society are incalculable, concerning, for instance, business opportunities, investment risks and government responses to crises. These externalities are complex externalities in addition to the ones which governments are supposed to address. Because clientelism is embedded in the way governments would interfere to tackle perceived market failures and market externalities, including complex risks such as climate change, pandemics and environmental degradation, it is conceptually difficult to disentangle market failure from government failure.

In conclusion, if normative theory must consider real-life situations and practices as feasibility considerations for the arguments it posits, just as the framework of constitutional political economy did, then clientelism is a useful addition to understanding the full range of constrains and limitations in actual politics beyond those which Buchanan and Tullock had in their mind when they made their own conceptual contribution. Taking this framework of normative evaluations closer to real-life circumstances can change our conceptual thinking and normative expectations about what the formal rules of the game can achieve. This paper challenges our convictions regarding how effective the formal rules of the game can be in shaping political outcomes knowing that informal exchanges will continue to shape these outcomes against these rules. It also points to the limits of constitutional reform given embedded incentives for these system of exchanges to linger on. A political system can hardly eradicate these exchanges by way of institutional reform even if most members of the political society publicly agree that this type of exchange is a harmful and illegitimate practice. It is ultimately about how the circumstances of politics affect markets and government simultaneously and in combination.