INTRODUCTION

As the film Yojimbo opens, director Akira Kurosawa shows matinee idol Toshiro Mifune ambling past a ramshackle hut. A young man, apparently the son and heir apparent, dashes out of the cottage. Mifune watches bemused as the son yells at his father. Go on and eat your rice gruel, he shouts. My life may be shorter than yours, but it won't be half as dull. The father is unimpressed. Get back to farming, he replies. The life of a swordsman is not fit for you.

Nor, Kurosawa seems to tell us, is the life of a swordsman fit for many of us – or of much use to any of us. For at least one of the themes in Yojimbo is the Hobbesian inefficiency of it all. The mythic world of Yojimbo is a world where no public authority enforces private claims to any scarce resource, and no private parties cooperate without that authority. Instead, those with money dissipate their fortunes by hiring Mifune and his fellow mercenaries to protect their wealth and rob their neighbors.

Were the farmers and merchants in Yojimbo willing and able to cooperate, they could agree to respect each other's claims and fire these private armies. But agreements to cooperate often presuppose a public mechanism to enforce them. Notwithstanding game-theoretic models of the evolution of cooperation or of self-enforcing contracts, the merchants in Yojimbo defect whenever it pays – and it pays often.

Jamaica tried several approaches to telecommunications regulations and experimented with both public and private ownership before settling in the late 1980s on private ownership and a new regulatory arrangement that seems to promote reasonably good performance in the sector and to be compatible with the country's institutions. Yet there is widespread public controversy about the current regulatory framework: a tight monopoly over all telecommunications services (including equipment supply), very little administrative discretion, and continuous price adjustments to satisfy what many consider too high a rate-of-return requirement. Should Jamaica change its regulatory system to one with greater institutional flexibility and fewer of the normative inefficiencies of rate-of-return regulation? Or, as the framework presented in Chapter 1 suggests, do Jamaica's institutional endowments explain the need for a system of rigid regulatory governance?

A traditional parliamentary government, strong judiciary, and strong two-party system, in which power is regularly transferred between the two parties, constrain the type of regulatory governance structure that can work well in Jamaica. A regulatory mechanism embedded in legislation of the U.S. style is too flexible and uncertain to provide the required safeguards for investment and growth in a system in which the rules of the game can change along with the government administration – or even within the same administration. If the courts rule that a particular administrative decision violates the regulatory statute, the government can overturn the ruling by passing new legislation.

The primary justification offered for the package vote and the confidence vote procedure by the drafters of the Fifth Republic Constitution was the need to stabilize the government by firmly entrenching executive authority over policymaking processes. As noted in the Introduction, research since 1958 has concluded that these two procedures have been successful to this end. The package vote and the confidence vote procedure are said to have effectively abolished the parliament's right of amendment, thereby ensuring that the government can implement its policy wishes when faced with a recalcitrant parliament. France's two restrictive procedures are thus viewed as institutional arrangements that influence the vertical relationship between the executive and legislature, giving the executive the upper hand when policy conflict with the National Assembly is severe.

There are two problems with the argument that the package vote and the confidence vote procedure are used by the government against the National Assembly for policy purposes. One problem is empirical. There is no reliable empirical evidence demonstrating that restrictive procedures are indeed used by the government in response to policy conflict with parliament. Instead, claims about policy conflict between the government and the National Assembly seem to be based on impressionistic evidence–on observation of the rhetoric in parliament that inevitably surrounds the utilization of the restrictive procedures.

The second problem with the prevailing view is that there exists no logical argument explaining why policy conflict between the government and the National Assembly should lead to the use of restrictive procedures.

The Cambridge series on the Political Economy of Institutions and Decisions is built around attempts to answer two central questions: How do institutions evolve in response to individual incentives, strategies, and choices, and how do institutions affect the performance of political and economic systems? The scope of the series is comparative and historical rather than international or specifically American, and the focus is positive rather than normative.

John Huber's original, imaginative book develops and tests theories about how institutional arrangements shape political decision making. He provides innovative game-theoretic arguments about how legislative procedures, through their impact on the bargaining strategies of political parties, influence policy outcomes, cabinet stability, and political accountability in France. Bringing together often-separated models of strategic bargaining and legislative behavior, he argues that most policy conflict in parliamentary systems is not between the cabinet and parliament but between different parties or factions within the government majority or between government and opposition parties during minority governments. He examines two institutional arrangements for resolving such conflicts, the ‘package vote’ and the ‘confidence vote procedure,’ both significant elements of politics and policy outcomes in the French National Assembly. He concludes that such procedures are less ‘antidemocratic’ than is often suggested. Neither the package vote nor the confidence vote procedure is used by the government to run roughshod over the National Assembly. Instead, the package vote is used by party leaders in government to preserve essential elements of policy bargains struck outside the legislature.

The United Kingdom has been a pacesetter in institutional change in the telecommunications sector. Over the 1980s the government privatized British Telecom, introduced a novel regulatory scheme and a new regulatory institution (Oftel, the office of Telecommunications), and opened the sector to competition in the network, customer premises equipment, and value-added network services. Private investors have shown remarkable confidence in the future of the sector despite the uncertainty generated by the lack of modern experience with regulation of private utilities, the intrinsic discretionary powers of the government in administrative decisions, and continuing institutional change.

The British political system provides no constitutional protection against discretionary regulatory behavior. The party in power controls both Parliament and the government, and there is no tradition of active judicial oversight of regulatory bodies. Thus governments and regulators cannot easily and credibly commit not to use administrative discretion to tighten the regulatory screws to expropriate a regulated firm's specific assets. Even if the courts rejected a particular regulatory interference, the government could get its way just by introducing new legislation or procedures. The puzzle, then, is how the Conservative governments of the 1980s were able to privatize the telecommunications, electricity, water, gas, and airport sectors. Why were private investors willing to invest large amounts in sectors that, in principle, were vulnerable to confiscatory regulation in the future?

At least part of the explanation lies in the way the privatization was handled and in the evolution and adaptation of British political institutions.

The Cambridge Series in the Political Economy of Institutions and Decisions is built around attempts to answer two central questions: How do institutions evolve in response to individual incentives, strategies, and choices; and how do institutions affect the performance of political and economic systems? The scope of the series is comparative and historical rather than international or specifically American, and the focus is positive rather than normative.

This volume, an outcome of a research project initiated by the World Bank, explores regulation of telecommunications in five countries. The editors' objective was to make a comparative assessment of the impact of core political and social institutions on regulatory structures and performance in order to derive some generalizations about what works and doesn't work in telecommunications regulation. The study is rich in original insights into the interplay between specific institutional endowments of a polity, governance structures, design characteristics, and consequent industry performance. More than anything else the study demonstrates the usefulness of the new institutional economics as a tool to analyze regulatory structures and aid in the development of sound regulatory practice.

Institutional arrangements in democratic systems do not magically materialize from an ethereal haze. Instead, they generally emerge from hotly contested debates among individuals who often will play a part in choosing future public policies. But why do political decisionmakers often exhibit such a deep interest in what often seem like trivial details of political rules and procedures? How do individuals who choose decision rules develop beliefs about the impact of specific rules on behavior? Are there systematic elements that underlie explanations of the choice of rules, or must any such ‘explanation’ be idiosyncratic and historically contingent?

This chapter addresses these questions by examining the choice of institutional arrangements through two quite different lenses. The first fixes on formal models of majority rule, and in particular on the contributions these models make toward establishing a general framework for thinking about the adoption of democratic institutions. The second lens fixes on the historical events in France in 1958 that led to the inclusion of the package vote and the confidence vote procedure in the Fifth Republic Constitution.

My goal is to create a dialog in two directions. On one hand, examining the specific events in France aims, not only to provide new historical details about the process that led to the choice of particular procedures in 1958, but also to highlight issues that are relevant to building general theories of endogenous institutions.

There are many different types of institutional arrangements that constitutional engineers can select to influence how political parties resolve conflict. The chief architect of the Fifth Republic, Michel Debré, wanted to select institutional arrangements that would be particularly suited to ending the extraordinary cabinet instability of the Fourth Republic. To this end, he advocated a single-member district, plurality rule electoral law. This institutional arrangement, he believed, would lead to stable governments by creating homogenous, single-party majorities at election time.

French elites ultimately declined this path, however, not because they objected to using the electoral law to manufacture single-party majorities, but because General de Gaulle feared that specifying any electoral law could jeopardize ratification of the Constitution by the French people. The framers therefore chose an alternative institutional solution to the bargaining problems of the Fourth Republic. They placed in the Constitution legislative procedures that would ensure governmental control over policymaking and that would weaken opportunities for destabilizing electoral competition on the floor of parliament.

The package vote and the confidence vote procedure were among the most controversial of the new constitutional procedures. Since each of these institutional innovations permits a member of the government to make the final, take-it-or-leave-it policy proposal, deputies can do little to evade the effects of these provisions. Visceral rhetoric therefore often surrounds discussions of the package vote and the confidence vote procedure, both of which are frequently denounced as antidemocratic because they effectively eliminate amendment prerogatives by deputies.

Two stylized facts are woven through the literature on Fifth Republic France. The first is the primacy of the president in French politics. Institutional arrangements, and in particular the direct election of the president since 1962, are said to give the president vast power. Presidential power is held to be further augmented by the presidentialization of French political parties: since the presidency is an enormous prize, parties are organized to service the needs of their presidential aspirants, and these présidentiables are believed to exercise extraordinary control over party positions on policy.

The second stylized fact is that the National Assembly plays no important role in policymaking. A whole host of constitutional procedures that ‘rationalize’ the role of parliament, together with the emergence of coherent legislative majorities under the Fifth Republic, are said to limit substantially the opportunities for France's elected representatives to participate meaningfully in the legislative process. The National Assembly is therefore often held to be among the weakest legislatures in the advanced industrialized world.

These two stylized facts have had a significant impact on political science research. On one hand, the presidential interpretation of the French system has often led comparative research on parliamentary government to exclude the Fifth Republic. A survey of nineteen studies of coalition formation, for example, reveals only five that include the Fifth Republic, while eight exclude France altogether and six focus only on the Fourth Republic.

People the world over are calling for privatization and regulatory reform as the solution to the problem of poor performance by public utilities, and as the means to improved service and lower prices. In this book we argue that these expectations may not always be met because of the way a country's political and social institutions – its executive, legislative, its judicial systems, its informal norms of public behavior – interact with regulatory processes and economic conditions. By making administrative expropriation and manipulation of utilities more or less difficult in some settings than in others, a country's institutions influence the confidence of investors and the performance of privatized utilities. In particular, the analysis in this book looks at the problem of utilities regulation through the lens of the new institutional economics – with its microanalytical perspective and its view of regulation as a contracting problem – to offer insights into the performance of privatized telecommunications utilities in different political and social circumstances.

The book presents the results of a comparative assessment of the impact of core political and social institutions on regulatory structures and performance in the telecommunications industry in Jamaica, the United Kingdom, Chile, Argentina, and the Philippines. We argue that the credibility and effectiveness of a regulatory framework, and so its ability to encourage private investment and support efficiency in the production and use of services, vary with a country's political and social institutions.

Chile's experiments with regulatory regimes and ownership patterns in telecommunications have produced radically different results. Between 1930 and 1970 firms were privately owned and regulated but the sector grew only modestly after the agreed investment obligations were fulfilled. In the 1970s the government took over ownership of the two firms providing local and long-distance services and ran them as regulated monopolies. The sector grew even slower than before. Beginning in 1982 Chile deregulated some segments of the sector, improved regulation of others, and returned the firms to private ownership. The gains have been substantial: The number of new lines doubled in four years, unmet demand has declined, new services are being introduced, and modern technology is replacing old. Why has this latest episode been so much more successful than the others?

Chile's experience shows how the proper match of regulatory regime and political institutions can foster efficient development of the telecommunications sector. Though regulatory incentive structures have been important (pricing, entry), more important have been the distributional demands of various interest groups and the effect of institutional factors on the credibility of government commitment not to behave opportunistically.

Three questions guide the analysis:

• How clearly did the different regulatory regimes spell out pricing rules, entry policy, and conflict-resolution mechanisms?

• Did these different regulatory regimes equally persuade firms to invest in the sector and operate efficiently?

• […]

Policymakers the world over have come to see privatization as an important remedy for the ills afflicting economies long dominated by stateowned enterprises. Private sector involvement is expected to bring increased resources and efficiency to ailing public enterprises. Yet it is still an open question whether private enterprises can perform well in situations where public ones have failed. Indeed, many private enterprises in developing countries have track records no better than those of typical public enterprises in similar situations. What are necessary conditions for the success of the private sector in a developing economy? Certainly, it takes more than a simple transfer of ownership for a firm or industry to prosper.

The experience of the telecommunications sector in the Philippines provides an opportunity to study the underlying causes of success and failure in an infrastructure sector in which most firms have been privately owned since the industry's infancy. The main operators in the sector have had direct access to international capital markets and have managed to raise large amounts of debt capital relative to their equity, yet the sector has not performed very well: telephone density is low (about 1.7 telephone sets and 1.1 main lines per 100 people), waiting lists for basic services are long, and service quality is roundly attacked (Gavino 1992). Even taking the country's level of development into account, it seems that the sector has had no edge over its counterparts in other countries, including those under public ownership or with more restricted access to foreign capital.

The new institutional economics is a way of reasoning and approaching political economic problems. Its objective is to broaden out and modify the microeconomic foundation of economic theory by taking into account the way in which institutions, political and economic, affect the performance of economics over time. It provides a far richer context because it explores the many other dimensions besides just price and quantity, the basis for the comparative model of economic theory that shape the performance of economics through time.

While the new institutional economics has developed a body of theory to analyze the way in which political and economic institutions evolve, it has been short on empirical content. This study is a major contribution to filling in that lacuna. It is a careful and thoughtful study of the institutional features of regulation over telecommunications in five different countries, and draws from that empirical material important implications for developing sound regulatory policy that will improve the performance of that industry.

The study makes clear that, in attempting to provide a regulatory framework for an industry of such dynamic technological change as that of telecommunications, intelligent regulatory policy must take into account the institutional foundations and endowments of a nation, and the way those can be extended and applied. The study speaks for itself in demonstrating unequivocally the power of the new institutional economics to provide a far deeper foundation for economic policymaking, analyzing, and modeling then we have had heretofore.

This chapter develops an Electoral Politics Model of the confidence vote procedure. Unlike agents in the models discussed previously, the agents in the Electoral Politics Model are not motivated exclusively by short-term policy preferences. Instead, they are also motivated by ‘office’ considerations (related to control of the government) and electoral considerations (related to communication with voters). The analysis explores how these various motivations influence strategic behavior in policymaking after government formation is complete. The result is a theory that directs attention away from policy considerations in procedural choice and toward the role that the confidence vote procedure plays in shaping opportunities for political parties to communicate to voters information about political accountability and policy positions.

The chapter has three sections. The first section describes and defends the assumptions of the Electoral Politics Model. I then sketch the results, and the logic underlying these results, in the second section. The third section concludes with a discussion of the empirical implications from the model.

THE MODEL: PARTY MOTIVATIONS AND CONFIDENCE VOTES

The basic structure of the Electoral Politics Model is quite similar to that of the Policy Conflict Model described in Chapter 3. But the Electoral Politics Model contains assumptions about the identity of the agents, their motivations (or utility functions), and their possible strategies (or actions) that are quite different than those found in the Policy Conflict Model. I review these differences in this section.

Democratic government is impossible without formal institutional arrangements, or rules, that fix the limits of legitimate political behavior. Rules define how elected representatives are chosen, how policies are formulated and adopted, how policies are enforced, and even how the rules for choosing policy are themselves established. Although there is an enormous variety of institutional arrangements for democratic government, almost all share one common feature: they are not neutral. Most rules create opportunities for particular individuals or groups, impose constraints on others, and thus affect who wins and who loses the competition to influence policy outcomes. Rules also create trade-offs. Some may increase the efficiency of decision making, but only at the cost of excluding certain individuals or groups from power and influence. Other rules may enhance political stability, but only by denying meaningful choices to citizens at election time. And yet other rules may ensure the inclusion of a large number of groups in decision making, but at the cost of clear accountability for policy choices. Not surprisingly, then, political elites expend enormous energies battling over the rules for political decision making.

The transition from the Fourth to the Fifth Republic in France provides what may be the most dramatic historical example of how changing the rules of a democracy can change the performance of that democracy. The Fourth Republic is often criticized for two related reasons. The first is executive impotence.