Keywords

JEL Classifications

Formal and informal institutions arise and evolve to underpin economic activity and exchange by protecting property rights, enforcing contracts, and collectively providing physical and organizational infrastructure. The field of economic governance studies and compares these institutions: state politico-legal institutions, private ordering within the law (credible contracting, arbitration), for-profit governance (credit-rating agencies, organized crime), and social networks and norms. Private institutions can outperform the state’s legal system in obtaining and interpreting relevant information, and imposing social sanctions on the violators of norms. But private institutions are often limited in size; as economic activity expands, a transition towards more formal institutions is usually observed.

Concepts and Taxonomies

The term ‘governance’ has exploded from obscurity to ubiquity in economics since the 1970s. A search of the EconLit database shows clear evidence of this explosion. In the relevant categories (title, keywords and abstracts), there are just five occurrences of the word from 1970 to 1979. The number jumps to 112 for the 1980s and 3,825 for the 1990s. Since 2000 to the time of this writing (December 2005), there are already 7,948.

The Oxford English Dictionary gives several definitions of the word ‘governance’: (a) the action or manner of governing; controlling, directing, or regulating influence; control, sway, mastery; the state of being governed; good order; (b) the office, function, or power of governing; authority or permission to govern; that which governs; (c) the manner in which something is governed or regulated; method of management, system of regulations; a rule of practice, a discipline; and (d) the conduct of life or business; mode of living; behaviour, demeanour; discreet or virtuous behaviour; wise self-command. These diverse meanings allow the word to be used (and sometimes misused) for almost any context of economic decision-making or policy.

Two areas of application merit special mention. One is corporate governance. This analyses the internal management of a corporation – organizational structure and the design of incentives for managers and workers – and the rules and procedures by which the corporation deals with its shareholders and other stakeholders.

The second is economic governance; Williamson (2005) expresses its theme as the ‘study of good order and workable arrangements’. This includes the institutions and organizations that underpin economic transactions by protecting property rights, enforcing contracts, and organizing collective action to provide the infrastructure of rules, regulations, and information that are needed to lend feasibility or workability to the interactions among different economic actors, individual and corporate. Different economies at different times have used different institutions to perform these functions, with different degrees of success. The field of economic governance studies and compares these different institutions. It includes theoretical models and empirical and case studies of the performance of different institutions under different circumstances, of how they relate to each other, of how they evolve over time, and of whether and how transitions from one to another occur as the nature and scope of economic activity and its institutional requirements change.

Corporate governance and economic governance are connected because the boundary of a corporation is itself endogenous, determined by the same considerations of information and commitment costs that raise problems of internal organization as well as those of property and contract (Coase 1937). Specifically, the nature of transaction costs may make it more efficient to handle some problems of governance by merging the two parties, for example by vertical integration (Williamson 1975, 1995). But it is analytically convenient to separate the two. This article concerns economic governance. To avoid constant repetition, I will simply call it ‘governance’ here unless some explicit reference to corporate governance is relevant.

Governance was neglected by economists for a long time, perhaps because they expected the government to provide it efficiently. However, experience with less developed and reforming economies, and observations from economic history, have led economists to study non-governmental institutions of governance.

Governance is not a field per se; it is an organizing or encompassing concept that bears on issues in many fields, including institutions and organizational behaviour, economic development and growth, industrial organization, law and economics, political economy, comparative economic systems, and various subfields of these.

We can organize the subject by classifying institutions along different dimensions. As is usual with such taxonomies, these are conceptual categories to help organize our thinking and analysis. In reality, there are significant differences within each category and overlaps across categories.

The first dimension concerns the purpose of the institution. The categories are: (a) protection of property rights against theft by other individuals and usurpation by the state itself or its agents, (b) enforcement of voluntary contracts among individuals, and (c) provision of the physical and regulatory infrastructure to facilitate economic activity and the functioning of the first two categories of institutions. We might also consider a fourth category, namely, the deep institutions that are essential to avoid serious cleavages or alienation that threaten the cohesion of the society itself. But this has not been studied in this context so far.

The second dimension concerns the nature of the institution. The categories are: (a) the formal state institutions that enact and enforce the laws, including the legislature, police, judiciary and regulatory, agencies, (b) institutions of private ordering that function under the umbrella of state law, for example various forums for arbitration, (c) private for-profit institutions that provide information and enforcement, and (d) self-enforcement within social or ethnic groups and network. My discussion is organized in sections along this dimension.

A third dimension distinguishes institutions that arise and evolve organically from those that are designed purposively; self-enforcing groups are often organic while the first three categories in the second dimension usually require some measure of design. This matters for the evolution of institutions of governance (see Greif 2006, especially ch. 6; Williamson 2005 p. 1).

Formal Institutions of the State

There is broad agreement that the quality of institutions of governance significantly affects economic outcomes. The importance of protecting property rights, both from other individuals and from predation by the state itself, is generally recognized and documented (for example, De Soto 2000). But serious disputes about the precise measures of quality of institutions, and about many details of the causal mechanisms by which they affect economic outcomes, remain.

At the broadest level, the distinction is between democracy and authoritarianism, each of which comes in many different varieties. Democracy has many normative virtues, but its worth in governance is less clear. Barro (1999, p. 61) finds an inverse U-shaped relationship between economic growth and a continuous measure of democracy – ‘more democracy raises growth when political freedoms are weak, but depresses growth when a moderate amount of freedom is already established’ – but the fit is relatively poor. Persson (2005), using cross-sectional as well as panel data, finds that the crude distinction between democratic and non-democratic forms of government is not enough. The precise form of democracy matters for policy design and economic outcomes: ‘parliamentary, proportional, and permanent democracies seem to foster the adoption of more growth-promoting structural policies, whereas ... presidential, majoritarian, and temporary democracy do not’ (Persson 2005, p. 22). However, Keefer (2004, p. 10), after surveying a wide-ranging literature on electoral rules and legislative organizations, concludes that they affect policies but are not a crucial determinant of success: ‘electoral rules ... almost surely do not explain why some countries grow and others do not’, and ‘the mere fact that developing countries are more likely to have presidential forms of government is unlikely to be a key factor to explain slow development.’

Democracy can be important for governance because its reliance on rules and procedures provides citizens with protection against predation by the state or its agents. Indeed, the elite, which might otherwise prefer to rule unconstrained, may find it in its own interest to make a credible commitment not to steal from the population by creating and fostering democracy (Acemoglu 2003; Acemoglu and Robinson 2005). Greif et al. (1994) discuss how groups of traders (guilds) in late medieval Europe took collective action to counter rulers’ incentives to violate their members’ property rights.

Even in a democracy, agents of the state may pursue their private interests using corruption, complex regulations to extract rent, and favouritism. In fact, an emerging literature argues that economic growth, at least in its early stages, is better promoted under suitably authoritarian regimes. Glaeser et al. (2004) argue that less developed countries that achieve economic success do so by pursuing good policies, often under dictatorships, and only then do they democratize. While these conclusions are controversial, these authors’ criticisms of the measures of institutions used in the research that argues for the primacy of institutions in general, and of democracy in particular, are telling. Giavazzi and Tabellini (2005) find a positive feedback between economic and political reform, but they also find that the sequence of reforms matters, and countries that implement economic liberalization first and then democratize do much better in most dimensions than those that follow the opposite route. In practice, of course, it is difficult to ensure ex ante that an authoritarian ruler will implement good governance.

Many different measures of institutional quality exist. World Bank researchers Kaufman et al. (2005, which contains citations to their earlier work) have constructed six: (a) Voice and Accountability – measuring political, civil and human rights; (b) Political Instability and Violence – measuring the likelihood of violent threats to, or changes in, government, including terrorism; (c) Government Effectiveness – measuring the competence of the bureaucracy and the quality of public service delivery; (d) Regulatory Burden – measuring the incidence of market-unfriendly policies; (e) Rule of Law – measuring the quality of contract enforcement, the police, and the courts, as well as the likelihood of crime and violence; and (f) Control of Corruption – measuring the exercise of public power for private gain, including both petty and grand corruption and state capture. Of these, (e), (f) and also (b) concern the most basic institutions for protection of property rights and enforcement of contracts, (a) relates to governance because voice and accountability can reduce the severity of the agency problem between the citizens and the agencies of the state, and (c) and (d) pertain to what I called provision of the infrastructure of governance. Conceptually they are a mixed bag; the quality of some of them can itself depend on the quality of other more basic ones, and some are closer to being measures of effects than of causes. Their method of construction relies on subjective perceptions, and is subject to error. But when used with caution, they have proved significant as explanatory variables in empirical studies of economic growth, and for observing changes in governance quality over time in specific countries. Corruption and regulatory burdens are major themes of the World Bank’s research on governance in many countries (see World Bank Institute, website).

Empirical estimations of the level or growth of GDP on various measures of institutional quality confront many conceptual and econometric problems. Researchers have tackled the issue of reverse causation by using various instruments, such as the nationality of colonizers (Hall and Jones 1999), mortality among colonizers (Acemoglu et al. 2001), and whether a colony had rich mineral resources or climatic and soil conditions conducive to plantation agriculture and a large or dense native population, or was sparsely populated and poor in the 1500 s (Engerman and Sokoloff 2002; Acemoglu et al. 2002). The general idea is that in the former circumstances the European colonizers established institutions of slavery and inequality to facilitate the exploitation of labour on a large scale, whereas in the latter conditions, where the colonizers had to exert their own effort, their institutions provided the correct incentives and became conducive to longer-term economic success. The debate on the factual and econometric validity and the economic interpretation of these findings is fierce and continuing; Hoff (2003) surveys and discusses this literature in detail.

La Porta et al. (1998, 1999) contrast different legal traditions for protecting the rights of small shareholders. If such protection is poor, that will inhibit the flows of capital to its most efficient uses. They find that systems based on common law are better in this regard than those based on civil law. But Rajan and Zingales (2003) and Lamoreaux and Rosenthal (2005) argue that in practice there was little difference between the systems during critical periods of industrialization.

These debates are sure to continue, and this section will get out of date very quickly.

At the international level, formal governance works through bodies like the World Trade Organization. Their members are sovereign countries; therefore their procedures must be subject to self-enforcement in repeated interactions, whether through bilateral or multilateral sanctions. These institutions are therefore basically similar to the social networks discussed below. See Maggi (1999) and Bagwell and Staiger (2003) for detailed analyses.

Private Institutions

The policing functions for property right protection supplied by the state are often supplemented by private security systems that serve specific clients and purposes – firms employ or hire security personnel, gated communities and neighbourhoods have private (hired or volunteer) patrols. These generally merely supplement the functions of the police for their specific context and work cooperatively with the police, but the two may clash if the private security system goes beyond its permissible functions.

Private institutions of contract enforcement similarly coexist with formal law, and become essential when the latter is weak or nonexistent. Explicit or implicit private contractual arrangements are also important for assignment of property rights as a part of Coasean contracting for efficient outcomes. Therefore, analyses of private institutions often focus on the governance of contracts.

The basic problem of contract enforcement is control of opportunism. If one or both parties have to make transaction-specific investments, the other can attempt to secure a greater part of the benefit by reneging or demanding renegotiation. The prospect of this can jeopardize the potentially mutually beneficial deal in the first place. Williamson (1975, 1995) pioneered the analysis of this issue under the title of transaction cost economics.

Information constitutes a major source of advantage for private ordering over formal law. Enforcement of a contract in a court requires offering proof of misconduct by the other party in the dispute; the relevant information must be verifiable to outsiders. Therefore, formal contracts can stipulate actions by the parties conditional only on verifiable information. Other or more detailed information may be observable to the parties themselves, or can be inferred by specialist insiders to the industry, but cannot be verified to non-specialist judges or juries of the state’s legal system, or can be verified only at excessive cost.

The informational advantage of private ordering may be offset by a disadvantage in enforcement. Informal arrangements must be made to overcome each participant’s temptation to behave opportunistically at the others’ expense. Different methods of this kind underlie the various institutions of informal governance, and achieve different degrees of success. Some are able to exert coercion for immediate punishment of misbehaviour. Others create long-run costs, typically in the form of exclusion from future participation or worse future opportunities, to offset the short-run advantages of opportunism. This is the standard theory of self-enforcing cooperation in repeated Prisoner’s Dilemmas. The following sections discuss some of these alternatives.

Private Ordering with Formal Law in the Background

Perhaps the most remarkable thing about formal legal institutions and mechanisms for the enforcement of commercial contracts is how rarely they are actually used. Business transactions often do have underlying formal contracts, but when disputes arise recourse to the law is often the last resort. Other private alternatives are tried first; these include bilateral negotiation, arbitration by industry experts, and so on. Filing a suit in a formal court of law often signals the end of a business relationship. Most actual practice in business contracting is therefore better characterized as ‘private ordering under the shadow of the law’ (Macaulay 1963; Williamson 1995, pp. 95–100, 121–2).

If one of the parties to an ongoing informal relationship behaves opportunistically, the most common alternative is to fall back on a formal contract based on verifiable contingencies alone. Suppose an outcome based on a tacit understanding of what each party should do in any one exchange (including good-faith negotiation to adapt to changing circumstances) yields both of them higher payoffs than does a formal contract. Consider the implicit arrangement where, if one party deviates from the agreed course of action to its own advantage and to the detriment of the other, their future exchanges will be governed by the formal contract. This yields a subgame-perfect (credible) equilibrium of the repeated game if each party’s one-time gain from opportunism does not exceed the capitalized value of the future difference of payoffs between the tacit and the formal contracts. Williamson (2005, p. 2) expresses this well: ‘continuity can be put in jeopardy by defecting from the spirit of cooperation and reverting to the letter.’

When such relationship-based implicit contracting prevails, partial improvement in the formal system can worsen the outcome, due to a problem of the second-best. The partial improvement raises the payoffs the two parties could get from the fallback formal contract. This in turn reduces the future cost of a current deviation from the implicit contract or spirit of cooperation. It tightens the incentive-compatibility constraints, and therefore worsens what can be achieved by relational contracting (Baker et al. 1994; Dixit 2004, ch. 2).

Arbitration comes in two prominent forms. One is industry-specific, based on expert knowledge of insiders. More information is verifiable in such settings; therefore richer contracts specifying actions for more detailed contingencies become feasible. In many industries there is a large common-knowledge basis of custom and practice, which may even make it unnecessary to write down a contingent contract in great detail. Arbitration can also provide an opportunity for the parties to communicate and renegotiate adaptations to new circumstances. Formal legal systems often recognize these advantages of expert arbitration, and courts stand ready to enforce the decisions of arbitrators if the losing party tries to evade the sanctions. However, industry arbitrators often have severe sanctions at their own disposal; they can essentially drive the miscreant out of business, and even ostracize him or her from the social group of that business community. Examples of arbitration institutions include Bernstein’s (1992) classic study of the diamond industry. For further discussion and modelling, see Dixit (2004, ch. 2) and Williamson (2005, p. 14).

The other prominent forums of arbitration deal with international contracts (Dezalay and Garth 1996; Mattli 2001). There are several of these, specializing in different legal traditions. They lack direct power to enforce their decisions, but are backed by treaties that ensure enforcement by national courts. These forums do not have industry-specific knowledge, their processes can be slow and costly, and their decisions can be somewhat arbitrary. But parties in transnational transactions may prefer them to either country’s courts, suspecting that these will be biased in favour of their own nationals.

For-Profit Private Institutions

If the state is unwilling to protect certain kinds of property or enforce certain kinds of contracts (for example in illegal activities), or is unable to do so (for example in weak and failing states), or is itself predatory, then private institutions can emerge to perform these functions for a profit. Organized crime often fills the niches uncovered by the state. Gambetta (1993), Bandiera (2003) and others argue that the Mafia emerged in just such a situation to fill the vacuum of protection in late 19th-century Sicily. Landowners began to hire guards of former feudal lords, and even the toughest among bandits, to protect their property. Gambetta describes how the Mafia’s role expanded to providing contract enforcement in illegal or grey markets. Similarly, the Japanese Yakuza was instrumental in organizing markets at the end of the Second World War in August and September 1945 when the Japanese state had collapsed (Dower 1999, pp. 140–8), and mafias grew in Russia after the collapse of the Soviet regime (Varese 2001).

Gambetta (1993, p. 19) argues that this ‘business of protection’ is the core business of the Mafia. It may engage in other activities using in-house protection, but that is just downstream vertical integration – the opposite of upstream integration where an ordinary business firm has its in-house security department. A transaction-cost analysis of the internal organization of mafias, and of their vertical integration decisions, may provide an interesting link between economic governance and corporate governance. Another dimension in which the protection business can expand is extortion; although private protectors may be welcome when state protection has collapsed, ‘protectors, once enlisted, invariably overstay their welcome’ (Gambetta 1993, p. 198).

The Mafia can provide contract enforcement because, even though two traders may not have sufficiently frequent dealings with each other to achieve good outcomes in an ongoing bilateral relationship, each trader can be a regular customer of the enforcer. This converts multiple one-shot Prisoner’s Dilemma games among the whole group of traders into several bilateral repeated games of each trader with the enforcer. The intermediary can provide information (keeping track of previous contract violations and informing a customer of the history of a potential trading partner) and/or actual punishment if a customer’s trading partner violates their contract. The information role of the Mafioso is similar to that of credit rating agencies and Better Business Bureaus in the United States. Dixit (2004, ch. 4) constructs a model of such for-profit governance, and establishes the conditions for an equilibrium with for-profit private enforcement. These are lower bounds on the shares of the surplus that the customer and the Mafioso must have, so as to overcome the trader’s temptation to cheat and the Mafioso’ temptation to double-cross the customer. Milgrom et al. (1990) have a related and complementary model of private judges at medieval European trade fairs. They specify the game of each trade, and investigation in the event of cheating, in greater detail, but do not examine the issue of the judges’ honesty.

Group Enforcement Through Social Networks and Norms

Any institution of contract enforcement must solve three key problems: (a) detection of opportunistic deviations from the contractually stipulated behaviour, (b) preservation and dissemination of information about the histories of the participants’ behaviour, and (c) inflicting appropriate punishments to reduce future payoffs of any deviators. The first is often constrained by the available technology of monitoring, although institutions and regulations such as reporting requirements and auditing can improve the technology. The second and third problems are best resolved in bilateral ongoing relationships: each party has a natural incentive to detect and remember the other’s cheating, and can punish the other by breaking off the relationship. However, governance is often needed in groups each of whose members interacts frequently with someone else in the group, but not necessarily bilaterally with the same person every time. Now remembering and transmitting information about your current partner’s behaviour to others, and refusing a potentially beneficial deal because the counter-party has cheated someone else in the past, are privately costly activities and therefore require their own governance mechanisms.

Formal state institutions of governance can solve these problems by fiat; the legal system compels the whole group of traders to commit to good behaviour by subjecting themselves to detection and punishment if they cheat. A third-party supplier of information or enforcement serves similar functions. In the case of a Mafia enforcer, anyone who trades with a customer of the Mafioso subjects himself to the grim punishment if he cheats. In the case of a Better Business Bureau, a firm that joins the organization thereby gives hostage to its own good behaviour: if it misbehaves it will get a poor rating or blacklisting. Transactions vary in their characteristics; therefore we should expect the effectiveness of such reputation mechanisms to vary also, and should not expect universal success from any one.

An institution of social networks and norms can solve the problems of information and punishment in a decentralized manner. Each participant can transmit information about his or her current trading partner’s behaviour to others in the group to whom he or she is linked. And each can play his or her assigned part in punishment, typically by refusing to trade, if he or she gets matched with a potential partner who is known to have misbehaved in past dealings with others in the group. Incentives to transmit information or refuse potentially good trades can be established by a norm that regards refusal to do so as itself a punishable offence, as in Abreu’s (1986) penal codes for repeated games; see Calvert (1995a, b). Extrinsic incentives may even be unnecessary if people have sufficiently strong natural instincts to punish social cheaters, as found by Fehr and Gächter (2000).

Numerous empirical and case studies of governance based on social relations have been conducted; space constraints allow mention of only a few. Greif’s (1993) historical analysis of Maghribi traders’ system of communication and collective punishment is well known. So is Ostrom’s (1990) synthesis of the evidence on common-pool resource management; she emphasizes the importance of local knowledge and communication, of appropriately designed (generally graduated) punishments, and of incentives for individuals to perform their assigned roles and actions in the system. Fafchamps (2004) studies and compares many different market institutions in Africa; his work highlights the importance of designing systems appropriate to the conditions of each country or group. Ensminger (1992) describes a similarly rich complex of arrangements for trade and employment relationships among the Orma tribe of Kenya, and examines how formal institutions of property right enforcement including title registration can interact dysfunctionally with traditional arrangements based on family and tribal connections. Johnson et al. (2002) present and analyse findings from survey research in former socialist economies. Of particular interest are the links between evolving formal and informal governance. Even without a backup of courts, trust in bilateral relationships can build quickly in response to good experiences. New or transient customers are more likely to be offered credit if courts work better, but the effectiveness of courts becomes largely irrelevant for the functioning of established relationships. Casella and Rauch (2002) study the role of ethnic networks in international trade.

Li (2003) points out a key difference between the costs of operating such a system and those of formal governance. A relation-based system of networks and norms has low fixed costs, but high and rising marginal costs. Trading on a small scale naturally starts among the most closely connected people who have sufficiently good communication and common understanding to sustain honesty. No fixed costs need be incurred to establish any formal rules or mechanisms of enforcement. But as trade expands, potential partners added at the margin are almost by definition less well-connected, making it harder to communicate information with them and to ensure their participation in any punishments. By contrast, formal or rule-based governance has high fixed costs of setting up the legal system and the information mechanism, but once these are incurred, marginal costs of dealing with strangers are low. Therefore, relation-based governance is better for small groups and rule-based governance better for large groups. Greif’s (1994) comparison between the relation-based system of Maghribi traders and the formal institutions of Genoese traders supports this theory. Dixit (2004, ch. 3) constructs a formal model that compares relation-based and rule-based systems. This characterizes the maximum size of a self-enforcing group, and finds that, when the group exceeds this critical size, the maximum scope of sustainable honesty shrinks absolutely. The intuition is as follows. At the critical size, each trader is indifferent between honesty and cheating when dealing with the most distant person. When more traders are added, this weakens the communication between the previously marginal person and other almost equally distant ones, tipping the balance toward cheating.

Kranton (1996) models individuals who can either choose bilateral long-lived self-enforcing trading relationships or search for one-time trading partners in an anonymous market with external enforcement. The market thus provides the outside opportunity in the repeated game of bilateral trade. If more people trade in the anonymous market, it becomes thicker and offers better prospects for successful search. Then parties in bilateral relationships have better outside opportunities, which makes it harder to sustain tacit cooperation there, further increasing the relative attraction of the market. Therefore the system can have multiple equilibria – no one uses the market because no one else uses it, or everyone uses the market because everyone else does – and can get locked into a Pareto-inferior equilibrium.

Evolution and Transformation of Governance Institutions

A persistent theme in this survey has been that different governance institutions are optimal for different societies, for different kinds of economic activity, and at different times. Changes in underlying technologies of production, exchange and communication change the relative merits of different methods of governance. As the volume and scope of trade expand, formal institutions generally become superior to informal ones, but informal ones serve useful roles under the shadow of formal ones even in the most advanced economies and sectors. All this raises the question of whether we should expect institutions to adapt and evolve optimally.

Williamson’s famous ‘discriminating alignment hypothesis’ says that transactions, with their different attributes, align with institutions, with their different costs and competencies; see his recent exposition (2005, p. 6). This gives ground for optimism for synergistic evolution of the need for governance and the institutions that supply it. Others are less sanguine. North (1990) and others argue that institutional change is subject to long delays due to resistance by organized interests favouring the status quo, problems of coordinating collective action to bring about a discrete change in equilibrium, and so on. Dixit (2004, pp. 79–85) discusses some of these problems for transition from relation-based to rule-based contract enforcement. Eggertson (2005) gives a dramatic example of how institutions restricting fishing and requiring costly mutual insurance persisted in Iceland for centuries after they had become obstacles to good economic performance.

I believe that a balanced approach is needed, recognizing the tendency towards synergistic alignment but also the obstacles to its realization. The net outcome will depend on many specifics of each context. Understanding and predicting the process requires a combination of approaches: case-based and analytical, inductive and deductive. Greif (2006) discusses, develops and applies such methodologies using historical studies of trade in medieval Europe.

See Also