«Jon Thor Sturluson AKADEMISK AVHANDLING Som for avlaggande av ekonomie doktorsexamen vid Handelshogskolan i Stockholm framlaggs far offentlig ...»
Topics in the Industrial Organization
of Electricity Markets
Jon Thor Sturluson
Som for avlaggande av ekonomie doktorsexamen
vid Handelshogskolan i Stockholm framlaggs
far offentlig granskning
Fredagen den 23 maj 2003 kl13.15
i sal 750 Handelshagskolan
Topics in the Industrial Organization
of Electricity Markets
t~\ STOCKHOLM SCHOOL OF ECONOMICS
~~~p!l EFI, THE ECONOMIC RESEARCH INSTITUTE EFI Mission EFI, the Economic Research Institute at the Stockholm School of Economics, is a scientific institution which works independently of economic, political and sectional interests. It conducts theoretical and empirical research in management and economic sciences, including selected related disciplines. The Institute encourages and assists in the publication and distribution of its research findings and is also involved in the doctoral education at the Stockholm School of Economics.
EFI selects its projects based on the need for theoretical or practical development of a research domain, on methodological interests, and on the generality of a problem.
Research Organization The research activities are organized in twenty Research Centers within eight Research Areas.
Center Directors are professors at the Stockholm School of Economics.
ORGANIZATION AND MANAGEMENTManagement and Organisation; (A) Prof Sven-Erik Sjostrand Center for Ethics and Economics; (CEE) Adj Prof Hans de Geer Center for Entrepreneurship and Business Creation; (E) Prof Carin Holmquist Public Management; (F) Prof Nils Brunsson Infonnation Management; (I) Prof Mats Lundeberg Prof Jan Lowstedt Center for People and Organization; (PMO) Center for Innovation and Operations Management; (T) Prof Christer Karlsson
ECONOMIC PSYCHOLOGYCenter for Risk Research; (CFR) Prof Lennart Sjoberg Economic Psychology; (P) Prof Guje Sevon MARKETING Center for Consumer Marketing; (CCM) Acting Prof Magnus Soderlund Center for Information and Communication Research; (CIC) Adj Prof BertH Thomgren Marketing, Distribution and Industrial Prof Bjom Axelsson Dynamics; (D)
ACCOUNTING, CONTROL AND CORPORATE FINANCEAccounting and Managerial Finance; (B) Prof Lars Ostman Managerial Economics; (C) Prof Peter Jennergren FINANCE Finance; (FI) Prof Clas
Chairman ofthe Board: Prof Hakan Lindgren. Director: Associate Prof Bo Sellstedt.
A dress EFI, Box 6501, S-113 83 Stockholm, Sweden • Internet: www.hhs.se/efI!
Telephone: +46(0)8-736 ·90 00 • Fax: +46(0)8-31 62 70 • E-mail firstname.lastname@example.org Topics in the Industrial Organization of Electricity Markets
~~ STOCKHOLM SCHOOL OF ECONOMICS,\:f;' EFI, THE ECONOMIC RESEARCH INSTITUTE Dissertation for the Degree of Doctor of Philosophy, Ph.D.
Stockholm School of Economics, 2003 © EFI and the author ISBN NR 91-7258-614-1 Keywords Industrial Organization, Electricity markets, retailin~ switching costs, search costs, discrete choice, regulation, experimental economics, quantity precommitment.
Cover illustration by Sigmund, used with author's permission.
Printed by Elanders Gotab, Stockholm 2003
EFI, The Economic Research Institute Stockholm School of Economics POBox 6501, SE 1-113 83 Stockholm, Sweden www.hhs.se/efi To Anna Acknowledgements No one is an island.
The story of my graduate education is long and winding, and several people and events are responsible for the unpredictable turn of events. One of the primary reasons for choosing Stockholm, was my earlier involvement in the Nordic Energy Research Program, when I worked on my Masters thesis at the University of Iceland, under the guidance of Guomundur Magnusson and Friorik Mar Baldursson. I came to know the school and my future supervisor, professor Lars Bergman, through the program. lowe a lot to these three men. They are also to blame for getting me involved with the subject of electricity markets.
Several other people have contributed to these essays with constructive comments and enlightening discussions. Among those are: Sveinn Agnarsson, Eirik Amundssen, Torstein Bye, Chloe Le Coq (coauthor), Niclas Damsgaard (coauthor and office mate for many years), Richard Green, Magnus Johannesson, Tobias Lindqvist, Gylfi Magnusson, Marten Palme, S¢ren Leth Pedersen and Johan Stennek. Last but not least Tore Ellingsen, my co-advisor, who never seizes to amaze me with his lightning insights. He has had more impact on my work than he knows.
Everyone in the Nordic Energy Research Program - Energy and Society, has helped directly and indirectly at the many, but always fruitful, seminars and conferences over the years. I am also very grateful to the program for financial support.
I enjoyed my stay in Stockholm enormously. The school and the department of economics has provided a supreme environment for us students. I would like to thank all the staff I've interacted with over the years. In particular the administrators: Pirjo Furtenbach, Kerstin Niklasson, Ritva Kiviharju, Ingrid Nilsson and Britt-Marie Eisler. I would also like to thank everyone at IoES in Reykjavik, in particular Tryggvi Herbertsson, for the excellent accommodation at the University of Iceland where this work was finalized.
Last but not least, thanks to my family and in-laws for putting up with me. No one has contributed more to this dissertation, apart from myself perhaps, than my wife Anna.
I have you to thank for everything. Without you I am nothing.
Reykjavik, April 2003 Jon Thor Sturluson Contents Acknowledgements
This dissertation consists of four essays, all related to the field of Industrial Organization of electricity markets, in one way or another. It is far from being a concise overview of the field, but rather eclectic bits and pieces of a greater puzzle that's being studied by a large group of people.
The first three essays deal with issues concerning retail markets for electricity and similar goods. The first one is an empirical study and the other two apply game theory. The fourth essay is different in many respects, primarily methodologically, as it reports a laboratory experiment on a classical theme in Industrial Organization: price competition with capacity precommitment. Even though this final essay is more general than the rest, it is still relevant for electricity markets as it concerns the validity of models often used to analyze competition of electricity market.
Chapter 1 The first essay tries to explain the coexistence of three characteristics of the Swedish retail electricity market: It concerns a homogenous product, while price dispersion is high and consumers are reluctant to switch from incumbent suppliers to rivals. Consumers' reluctance to switch, as in this case, is usually explained either by search costs or switching costs, or even both. The purpose of the paper is to evaluate the relative importance of these two factors and study their interplay. The sequential choice problem facing consumers, a) whether to become an active searcher or not, and b) whether to switch suppliers or not, is modeled as two consecutive discrete choice problems and estimated using Swedish data.
I find that switching costs are, on average, larger than search costs. Yet, search costs are important for it appears that users of electricity must become active consunler before they respond to changes, in switching costs to any degree.
v Various forms of passive information such as direct marketing and word-ofmouth, encourage active searching.
Chapter 2 Essay 2 develops a simple model of a asymmetric price duopoly, where one firm is an incumbent with an established relationship to all consumers, while the other firm is a new entrant in the market. Consumers have different consumption levels and can face a switching cost or a search cost. Retail electricity markets are the prime example. Given conditions on the distribution of these costs, an asymmetric equilibrium in prices exists. In equilibrium, the rate of switching can be a bad measure of market performance. A decrease in switching costs affects prices for the benefit of all consumers and not only those who choose to switch. A decrease in search costs is likely to be primarily in the interest of those consumers that do not wish to search. This suggests that consumers may be facing a freerider problem, discouraging search as well as switching.
Chapter 3 In the third essay, Niclas Damsgaard and I consider an important policy issue concerning vertically integrated firms in network industries. An incumbent firm provides regulated local network services while being active in the deregulated but imperfectly competitive retail market. Its total costs are known, both to the regulator and its competitor, while the distribution between the two services is private knowledge of the incumbent. The firm has a natural incentive to overstate its costs for the network services to receive a higher regulated price.
However, since total costs of both operations are known, a claim of high network costs signals low retail costs. When firms compete in prices, which are strategic compliments, the integrated firm would like its competitor to belief that retail costs are high as well. The regulator, through a combination of price incentives and monitoring, can utilize this trade-off and induce an information-revealing separating equilibrium in which the incumbent claims its true type. The optimal combination of price incentives and monitoring and the conditions under which such a separating equilibrium is preferred to a pooling equilibrium are derived.
vi Chapter 4 The fourth essay, co-authored by Chloe Le Coq, investigates why subjects in laboratory experiments on capacity constrained price competition seem to consistently choose capacity above the Cournot level - the subgame-perfect equilibrium. We argue that this puzzling regularity may be attributed to players perceptions about their opponents skill or level of rationality. In our experimental design we used the level of experience (number of periods played) as a proxy for the level of rationality and matched subjects with different levels of experience.
We found evidence of capacity choices being decreasing and prices increasing with opponent's experience. This suggests that if subjects have a tendency to underestimate the rationality of their opponents, or that rationality is actually limited for a large proportion of subjects, the observed regularity need not be a puzzle after all.
Three important characteristics of the Swedish retail electricity market:
homogenous product, considerable price dispersion and low rate of switching from incumbent suppliers to rivals, are a cause of concern for competition policy. Consumer's reluctance to switch, as in this case, is usually explained either by search costs or switching costs, or even both. To evaluate the relative importance of these two factors the choices to, a) become an active searcher, and b) switch suppliers, are modelled as two consecutive discrete choice problems and estimated using Swedish data. We find that switching costs are, on average, larger thari search costs. Yet, search costs are important, for it appears that a great majority of users, of electricity need to be active consumer if they are to respond to changes in switching costs. Various forms of passive information such as direct marketing and word-of-mouth, encourage active searching.
*Several people have contributed to this paper with constructive comments on earlier stages.
Among those are: Sveinn Agnarsson, Niclas Damsgaard, Magnus Johannesson, Chloe Le Coq, Richard Green, Marten Palme and S¢ren Leth Pedersen. I would also like to thank participants in seminars at the Stockholm School of Economics, Institute of Economic Studies and on two occations within the Nordic Energy Research Program. A few people at the Swedish National Audit Office, where I obtained the data, deserve a fair share of thanks, especially Anders Berg.
Last but not least, I thank my supervisor, Lars Bergman for his guidance. The project was financed by the Nordic Energy Research Program.
1 Introduction In this new age of information consumers are constantly facing an increasing number of complicated choices. A surge in public use of information technology has lead to an incredible expansion of product information through advertising, direct marketing and new market mechanisms. At the same time, increased liberalization has created new markets where exchange was previously regulated, giving consumers opportunities to choose from a larger variety of products and suppliers. Important examples from the last decade are telecommunication services and energy, Le., electricity and gas.
Persistence in consumer choice is an important feature of many such markets and is observed in the low rate of switching from incumbent suppliers, such as prederegulation monopolists, to alternative rival service providers. Thue enough, the rate of switching can be a misleading indicator of market performance or competitiveness. It need not be a manifestation of market power. Consumers have, for instance, little reason to switch if markets are very conlpetitive and price dispersion low. Moreover, even if prices differ considerably between suppliers and brands, alarm bells should not go off automatically if product attributes other than price are important. Nevertheless, we can find clear examples of markets for fairly homogenous goods with low rate of consumer switching and high price dispersion. In such cases it is doubtful that competition is working as it is supposed to; for the benefit of consumers. Here we take a look at one such market where these three characteristics coexist - the retail electricity market in Sweden.