Services tailored to the utility of all

Nations, firms and citizens increasingly depend on the provision of high-quality, well-priced networks: telecommunications, electricity, transport and water. This in itself is enough to encourage active state involvement. The economic and legal features of networks, such as long-term capital investment requirements, externalities and natural monopoly elements, added to political, legal and cultural traditions of state action, provide further reasons.

State involvement can take many forms, hence the central regulatory issue is the institutional design of markets and the role of the state. Who should own essential networks - the state or private firms? Who should regulate them - national governments, subnational governments or specialised agencies at arm's length from elected politicians? What controls should be placed over regulators?

Essential networks in Western Europe were traditionally regulated through public ownership and much direct government control. In contrast, the US had mostly private ownership, together with limited subnational ownership and regulation through a mixture of independent regulatory commissions and elected politicians. However, such traditional arrangements have come under attack and have been altered over the past 30 years. The institutional design of network regulation has become politically salient owing to the technological and economic developments that have transformed industries such as telecommunications; new ideas about the problems of public ownership; pressures from overseas interests such as multinationals or the World Bank; and, more recently, experience of the benefits and problems of privatisation and independent regulatory agencies.

Two related questions arise. First, how to explain the way the regulation of networks has changed. Second, which institutional design works best.

José Gómez-Ibáñez examines both. He sees institutional design as a matter of transaction costs and contracts. Network infrastructures have specific features that create high transaction costs. They involve long-term investments, often with high initial costs. They are frequently, at least partially, natural monopolies. These features provide much scope for opportunism - by producers who may be monopoly producers of essential networks and have information advantages; by consumers who may fail to honour agreements to pay for assets already bought or that need replacing over the long term; and by governments that can manipulate prices, returns and ownership for short-term political gain.

Policy-makers can choose from a range of institutional designs. These divide into four types: public ownership; concessions to private companies; licences with discretionary regulation (usually via a specialised agency with some degree of independence from elected politicians that has discretion to respond to changing circumstances); and private contracts given to companies, regulated under general contract and competition law.

Gómez-Ibáñez uses a rich set of case studies to analyse how and why these different institutional designs are chosen and reformed. The author's range is exceptionally wide: Sri Lanka, the US, Britain and Argentina; he tackles telecommunications, railroads, airlines, electricity, buses and water; and spans time periods from the 1950s to today. The result is a remarkable array of examples, information and analyses. In explaining institutional design, he offers a series of fascinating arguments using opportunism and transaction-cost analysis as a starting point rather than as a constraining simplifying theme. Nevertheless, three broad features emerge.

First, institutional designs and reforms arise from many factors - economic, technological, political and legal. Therefore, multicausal and, indeed, multidisciplinary explanations are needed. Second, historical legacy is crucial: many choices are made in response to inherited problems and often in reaction to previous regulatory structures. As the flaws of each institutional design become evident, so criticisms mount and reforms are introduced, producing institutional "cycles". Third, choices arise from a fit between sectoral institutions and national ones such as federalism, political parties or the extent of regime stability. Thus sectors must be analysed in their context.

Which institutional design is the best? Each has advantages and disadvantages and offers powerful trade-offs. For instance, long-term concessions offer stability but pose difficulties if circumstances change, while discretionary regulation may allow vested interests to gain control of regulatory agencies. Gómez-Ibáñez begins by claiming he can effectively rank the four designs, with public ownership as the worst and private contracts as the best. However, the case studies offer a more complex picture, and this claim is the weakest point. Different forms of each institutional design are better, or more appropriate, for particular circumstances than others. For example, complex systems with many actors with diverse incentives are likely to fail. The privatisation of the British railways in the 1990s provides an excellent example of the disastrous effects of combining unbundling with other policies such as increasing supply, decreasing subsidies and creating several regulators.

Equally, the lowest ranked design - public ownership - can have positive effects, as occurred in the provision of buses in Sri Lanka (then Ceylon) in the late 1950s and 1960s. Indeed, it is a great pity that the author did not examine cases in France or Germany, which have successful publicly owned and operated railway and electricity infrastructures. Moreover, since each system weakens over time, it is likely that gradually the advantages of a new design are overwhelmed by increasing problems, so there is no one permanently best design. Finally, the author accords only a limited value to politics and decision-making processes: they feature as an explanatory factor, but not for their importance in aiding citizens to make better choices and to accept the outcomes of decision-making.

In contrast, Greg Palast, Jerrold Oppenheim and Theo MacGregor place politics and processes at the centre of their argument. They claim that regulatory processes that allow transparency and participation for users produce fairer outcomes. They offer biting criticisms of closed systems, such as price controls in British electricity, which, they argue, were largely negotiated between regulators and suppliers. Equally, they attack attempts to design competitive markets, as in California's electricity supply. Instead, they laud the virtues of what they claim is traditional US regulation: commissions that take evidence from all interested parties; cross-examination of suppliers; rights for utility users and employees to demand detailed evidence (including on sensitive matters such as costs); and open decision-making by regulators subject to challenge. They offer many examples of the dangers of closed decision-making, including the case of regulators agreeing over-lax price controls with suppliers, promptly followed by large increases in company stock-market valuations. They contrast these with instances in the US of processes that ensure discovery of information and stop unjustified price increases, layoffs of employees or restructuring of networks by suppliers. They also provide data to support the lower prices that open processes are claimed to provide.

The advantages of open decision-making seem attractive. Moreover, the book offers a welcome counterblast to assumptions about the superiority of competition, or the easy trust placed in specialised regulatory agencies that may fail to obtain sufficient information or allow real user participation. Its data on prices are valuable and should be systematically compared with figures such as those provided by Gómez-Ibáñez. Nevertheless, the book is rather polemical, suffers from repetition and lacks a theoretical model to underpin its claims. Also it makes little reference to existing academic work on utilities and is very US-centred, with some inclusion of the UK. Arguments would be strengthened by applying them to other nations.

These two works represent valuably contrasting arguments and deserve a wide readership among students of regulation of network industries, policy-makers and those working on regulation of markets. They show that infrastructure regulation remains politically and economically controversial, offering fascinating terrain for economists, political scientists and lawyers.

Mark Thatcher is senior lecturer in government and member, Centre for the Analysis of Risk and Regulation, London School of Economics.