Hold-up in Regulated Contracts: the Argentinean Natural Gas Transmission System Case
Abstract: This article aims to understand the role of policy stability perception in the dynamic of network infrastructures regulation. We contribute to the literature by developing an abstract description in which the regulatory institutions in some countries have virtuous relation with network industries, while other countries enter in a vicious cycle. The asset specificities inherent to network industries mean high transaction costs, which in turn raises the hold-up risk. We depart from the idea that regulation (as tariffs structures) is a kind of contract between government and private companies. As explained by Williamson (1976) and Goldberg (1976), it is a special kind of arrangement in the presence of incomplete contract that is able to protect players from holding up themselves. However, regulation can actual play a positive or a negative role in the network infrastructure development. We depart from the contract theory proposed by Salant and Woroch (1992), who model the governments’ incentives to behave opportunistically according to investors’ investment profiles. We show that their analysis, which is based on the incentive compatibility principles, explains behaviour differences if the investment profile of the industry is heterogeneous. However, it is no able to explain why industries with similar investment profile in different countries have completely different dynamics. Stein et al. (2008) underlined the importance of policy stability to understanding the Latin America success (or failure) in implementing policies. We include the variable policy stability perception as a significant element in the understanding of the government’s incentives to behave opportunistically. We check our model with the case study of natural gas network in Argentina. This analysis contributes to understand the role network industries regulation to deal with hold-up problem and how the institutional environment in which regulatory agencies are embedded matters.
