Estimating Size and Scope Economies in the Portuguese Water Sector Using the Most Appropriate Functional Form

Abstract

Contrary to the usual method of most studies that apply parametric methodologies, the current study does not assume a priori a functional form to represent the cost function. Instead, a large number of functional forms are tested in order to find the most appropriate functional form for the true cost structure of utilities. The sample focuses on water utilities operating in Portugal during the period 2002–2008. The results show that there are substantial economies of output density, economies of size, and economies of scope (in the joint supply of water supply and wastewater services) and, therefore, there are huge advantages in utilities merging.

Notes

The Appendix shows how the share equations formula associated with a general cost function were obtained.

The wholesale component encompasses the intake, treatment, and transportation in water supply and transportation and treatment in wastewater services and the retail component includes the water distribution and wastewater collection.

The capital letters D and W designate, respectively, the water supply and the wastewater and the lowercase letters r and w are related to the retail and wholesale components.

Figure 1 Observations categorized by year and according to the type of services provided.

It is said that these water utilities operate essentially in the retail component because some of these utilities are also responsible, totally or partially, for the wholesale component, in either the water supply or wastewater services. The water utilities, with their own bulk water production of more than 50% of the water provided, were also considered to be responsible for the wholesale component, as well as the utilities that treat more than 50% of the collected wastewater.

The total volume of water distributed (in the wholesale and retail) includes the entire amount of water that utilities abstract and/or acquire to the utilities providers (operating in wholesale) and includes the amount of water supplied to customers and the amount of water loss.

These dummies take the value 1 when the observations correspond to the respective type of utility and the value 0 in the remaining situations.

We should highlight that the specific EOD and Esize relative to output wastewater collection are negative because most utilities do not collect wastewater but only treat it. However, these results are not statistically significant.

Additional information

Notes on contributors

Pedro Carvalho

Pedro Carvalho is a B.Sci. and Ph.D., as well as a Researcher of Instituto Superior Técnico (IST) at the University of Lisbon, Portugal, and a research member of the Center for Urban and Regional Studies (CESUR). He holds a degree in civil engineering and his areas of interest include econometrics, search for economies of scale and economies of scope, benchmarking, and the performance measurement of public utilities.

Rui Cunha Marques

Rui Cunha Marques, B.Sc., M.Sc., and Ph.D., with a background in different fields (engineering, economics, and law), is Professor of Public Decision and Information at Instituto Superior Técnico of Technical University of Lisbon in Portugal. He is a research fellow at the Center for Urban and Regional System (CESUR) at the University of Lisbon and at the Public Utility Research Center (PURC) of the University of Florida in the United States and at the Centre of Local Government Studies of the University of New England in Australia. His areas of specialization include public utilities regulation, performance evaluation, project management, contracts and public procurement, particularly public–private partnerships (PPPs) and institutional issues, mainly in local government and public utilities and transportation, especially in the urban waste and water sectors.