Organized Crime
Theories of the Firm

With the notable early exceptions of Beccaria and Bentham, economic analysis of criminal activity came to the attention of the profession in the late 1960s. Becker interpreted individual criminal behaviour as the solution of a portfolio choice problem and studied the normative question of the optimal punishments to be used to enforce a given legislation. His main contribution was `to demonstrate that optimal policies to combat illegal behaviour are part of an optimal allocation of resources' and that the `economic framework becomes applicable to, and helps enrich, the analysis of illegal behaviour.' The 1970s and 1980s have seen substantial achievements in this literature, including further research in the microeconomics of criminal behaviour, the industrial economics of illegal markets and the optimum enforcement of laws. These issues formed the main focus of a joint conference with the Centro Interuniversitario di Studi Teorici per la Politica Economica (STEP) and the Università di Bologna, held in Bologna on 11/12 June, at a time of considerable turbulence in public order in Italy. The conference was organized by Stefano Zamagni, Professor of Economics and Head of Department at the Università di Bologna, Giorgio Basevi, Professor of Economics at the Università di Bologna and member of CEPR's Executive Committee, and Gianluca Fiorentini, Professor of Economics at the Università di Firenze.

Internal Organization of Criminal Firms
In her opening contribution, `Organized Crime, Mafia, and Governments', Annelise Anderson (Hoover Institution) applied a transaction costs approach to study the operations of the Sicilian Mafia. This allows more realistic assumptions about human nature (especially opportunism) than the neoclassical theory of the firm and emphasizes microeconomic details of its functioning that go beyond its role as a technological production function and focus instead on its role as a governance structure. Such a structure based on the `family' or cosca may be particularly suitable for enforcing and monitoring contracts in illegal markets, which are naturally characterized by low levels of trust. Pino Arlacchi (Università di Firenze) welcomed the ability of this approach to examine status relationships within criminal organizations and firms.

In `Rival Kleptocrats: The Mafia versus the State', Herschel Grossman (Brown University) presented a model of a kleptocratic State which attempts to maximize political rent by imposing taxes and providing public services. The Mafia competes in providing such services, and this competition may have both positive and negative effects: by causing the State to reduce taxation and increase provision of public services, it raises total production and hence the welfare of the representative producer, but socially disruptive Mafia activities that threaten the viability of the State may cause such benefits to be lost. To the extent that the Mafia forces the State to reduce taxation to maintain its subjects' allegiance, it causes a redistribution of national income from the political establishment to the citizenry. To the extent that the Mafia provides services to socially unproductive or counterproductive activities which the State does not support, however, such as theft, it causes national income to fall.

Marco Celentani (Universidad Carlos III, Madrid) commented that if the Mafia can benefit society by moderating the State's kleptocratic tendencies, an institutional reform to introduce competition among government agencies could dramatically reduce the distortions caused by a monopolistic government. Diego Gambetta (Department of Applied Social Studies, Oxford) criticized Grossman's assumption that property rights exist ex ante; in many areas where the presence of criminal organizations is large, it is the Mafia that settles and enforces such rights.

In `Gangs as Primitive States', joint with Constantinos Syropoulos, Stergios Skaperdas (University of California, Irvine) presented a model of `gangs' as long-lived organizations engaged primarily in criminal activities, each of which has a near-monopoly on violence within its own territory. These arise out of anarchy, in which those with comparative advantage in the use of force generally prevail; unlike other organizations examined by economists, such gangs emerge primarily through coercion rather than contract. The ideologies that rationalize gangs to their leaders, members and communities are correspondingly different, since they emphasize the importance of predatory behaviour as a determinant of success.

William Baumol (Princeton University) suggested extending the model in a dynamic context to establish the conditions leading to a successive impoverishment of the economy as disruptive activities increase and productive activities decline.

Industrial Organization
In `Oligopolistic Competition in Illegal Markets', Gianluca Fiorentini focused on criminal firms that produce an illegal commodity and also invest resources in violence and/or corruption to reduce the effectiveness of enforcement activities and hence also the costs of completing transactions in the output market. He considered models in which such firms act non-cooperatively, cooperate only in anti-enforcement activities, or cooperate in such activities and also in the output market. He found that the relationship between the level of resources invested in violence and corruption with collusion or competition is far from simple. It depends on how criminal firms expect enforcement agencies to react to an increase in such activities and the nature of the market in the illegal output. Equal investments in enforcement activities may generate quite different effects depending on the mechanism through which equilibrium is achieved in the output market (Bertrand or Cournot).

Samuel Peltzman (University of Chicago) questioned the passive role played by the police in Fiorentini's framework, in which the enforcement agency's responses are exogenously given. If the mutual effects of violence and police response were taken into account, different conlcusions would follow. Michele Polo (Università Bocconi, Milano) noted the importance of the assumption that investments are sunk, since there is substantial free entry in certain illegal markets such as drug trafficking.

In `The Reputational Penalty Firms Bear from Committing Criminal Fraud', written with Jonathan Karpoff, John Lott (University of Pennsylvania) used US data on 132 cases of alleged and actual corporate fraud during 1978-87 to show that an accused firm will suffer a significant reduction in its market value, of which only some 6.5% may be attributed to its expected legal fees and penalties; reputational effects therefore play a major role in disciplining firms that commit fraud. He related this to findings that parties who believe they have received products of lower quality than they were promised typically do greater damage to firms' reputations than do government regulators (whose intervention in product safety recalls for example may force customers to pay for quality assurance to which they attach little value). Correspondingly, frauds of stakeholders, government agencies and investors that are willing to press charges entail large reputational penalties, while such penalties for frauds that only concern regulatory violations and involve no `victims' are negligible.

Certain industries in Sicily and the US have persistently felt the influence of organized crime, usually in the form of enforced customer allocation agreements among independent firms who pay racketeer income for this service. In `Conspiracy Among the Many: The Mafia in Legitimate Industries', Diego Gambetta and Peter Reuter (Rand Corporation) considered the characteristics of the affected industries, the methods by which organized crime exercises its influence and the consequences of that influence. In a series of case-studies, they sought to identify the relationship between racketeer involvement and changes in technology and market characteristics. Low technology, slight product differentiation, inelastic demand and large numbers of suppliers appeared to facilitate racketeer involvement. Differences in racketeering between the US (predominantly the New York metropolitan area) and Sicily may reflect differences in the Mafia's organization: Mafia families have geographical territories in Sicily and functional territories in New York. US racketeering usually involves unions to disguise extortion as protection of workers' rights; in Sicily it more often flows from political corruption or the failure of the `market for trust'. The difficulty of eliminating organized crime from markets arises from a `reputational barrier to entry', which is not amenable to conventional interventions and which even successful prosecutions are likely to augment. Gambetta and Reuter concluded by examining some reform efforts that have sought to deal with the enforcement of cartels by criminal organizations.

Illegal Markets and State Intervention
In `Auditing with Ghosts', written with James Gordon, Frank Cowell (LSE) developed a model of tax evasion in which firms have different market opportunities and cost structures, and the tax-enforcement authority has access to various types of information. Each firm faces a competitive market for legal sales and a monopolistic market for illegal sales and must choose from three options: behaving honestly, diversifying between legal and illegal (unreported) sales, and becoming a `ghost'. This involves making no report rather than a false report to the tax authority. Cowell reported that the presence of ghosts reverses some of the conventional wisdom on enforcement strategies. Even an unsophisticated policy of random enforcement may yield more revenue in some cases than a `cut-off rule' whereby the tax authorities always (never) audit firms that report less (no less) than a certain turnover. In practice, some combination of both is likely to be optimal.

Vito Tanzi (IMF) noted that the tax authority can easily verify whether a given firm became a `ghost' by examining the files of subsequent fiscal years, which reduces the profitability of the evasion by increasing the risk of an ex post audit. Carla Marchese (Università di Torino) suggested raising existing penalties as an alternative means of fighting against `ghosts'.

The literature on the economics of crime has also focused on corruption. In `Corruption and Economic Growth', Isaac Ehrlich (State University of New York at Buffalo and Hong Kong University of Science and Technology) and Francis Lui (State University of New York at Buffalo) investigated whether corruption and growth are systematically correlated across different stages of development and possible causal mechanisms or specific factors to account for such a relationship. They developed a model in which human capital drives growth and government intervention creates a gap between free-market and `shadow' (government-imposed) prices, which inevitably leads to corruption. Bureaucratic position entails opportunities to obtain rents or bribes, which provides incentives to invest resources in the political capital that guarantees access to such position or power; these compete with incentives to invest in productive human capital.

Ehrlich and Lui considered the interplay between these two capital types and its consequences for growth and development under two institutional arrangements: either all workers and bureaucrats compete for access to political power or bureaucrats control investment in political power through an effective monopoly position. They drew lessons for the relationships between government intervention, corruption, and the pace and stability of economic growth. In particular, a command economy in transition to the free market with an efficient party organization that can assure the bureaucracy `organized corruption' may achieve steady-state growth even with government intervention. If party discipline disintegrates, however, leading to `disorganized corruption', any shock may move the economy off its equilibrium growth path into stagnation.

Pierluigi Sacco (Università di Firenze) criticized the authors' human capital production function; for economies at significantly lower levels of development than the industrialized countries, the return to investment in human capital depends strongly on the actual level of political capital, while the nature of corruption is closely related to the structure of society.

In `Corruption. Arm's Length Relationships and Markets', Vito Tanzi questioned the conventional assumptions that policy-makers and public sector employees promote the public interest and are fully compensated by legitimate wages and fringe benefits. Corruption invalidates the theoretical efficiency of public intervention, especially if it is widespread. It affects the role of the state through favourable treatment of particular taxpayers, discretionary application of regulations, and the granting of contracts for investments and other activities to particular individuals. The burden on taxpayers is therefore higher than the tax/GDP ratio suggests, the incidence of the tax system is affected, and the `cuts' taken by those making decisions reduce the effective level of public sector investment.

Flavio Delbono (Università di Bologna and CEPR) criticized Tanzi's assumption that perfect markets need no government intervention, which takes no account of the state's redistributive function. The possibility of implementing distortionary redistribution through corruption indicates a need for limited government intervention. Giovanni Dal Monte (Università di Napoli) stressed that the presence of arm's-length relationships may provide sources of development and of clientelism and corruption depending on the social context.

Governments seek to control the expansion of the criminal sector by minimizing aggregate profits to criminal firms from the sale of illegal goods, such as drugs or arms. In `Regulating the Organized Crime Sector', Marco Celentani (Universidad Carlos III, Madrid), Massimo Marrelli (Università di Napoli) and Riccardo Martina (Istituto Universitario Navale, Napoli) considered a repeated game with discounting among the government and two criminal firms to show that the government can devise a strategy that rewards criminal firms with a zero termination probability provided they conform to a low-profit (and possibly low-activity) profile. The government punishes a firm that deviates from such a path by concentrating all its resources on fighting it, which imposes a maximal termination probability. The authors showed that this strategy is feasible for a wide range of cases of strategic interaction. Finally, they presented a simple model of spatially differentiated criminal duopolists to show that the minimization of criminal firms' aggregate profits and minimization of the quantities of illegal goods traded can be made to coincide.

The Economics of Organized Crime edited by Gianluca Fiorentini and Sam Peltzman is a selection of papers presented at this conference.
ISBN (hardback) 0 521 47248 2
ISBN (paperback) 0 521 62955 1
Available from CEPR, 90-98 Goswell Road, London EC1V 7RR (fax: 44 20 7878 2999)