Vertical structure and the risk of rent extraction in the electricity industry

• Author: Anette Boom,Stefan Buehler
• Date: 01 January 2020
• DOI: http://doi.org/10.1111/jems.12327
• Published date: 01 January 2020

J Econ Manage Strat. 2020;29:210–237.wileyonlinelibrary.com/journal/jems210

|

© 2019 Wiley Periodicals, Inc.

Received: 12 November 2015

|

Revised: 4 June 2019

|

Accepted: 2 July 2019

DOI: 10.1111/jems.12327

INVITED REVIEW

Vertical structure and the risk of rent extraction in the

electricity industry

Anette Boom

1

|

Stefan Buehler

2

1

Department of Economics, Copenhagen

Business School, Frederiksberg, Denmark

2

Institute of Economics (FGN‐HSG),

University of St. Gallen, St. Gallen,

Switzerland

Correspondence

Anette Boom, Department of Economics,

Copenhagen Business School,

Porcelænshaven 16 A, DK‐2000

Frederiksberg, Denmark.

Email: ab.eco@cbs.dk

Funding information

Swiss National Science Foundation,

Grant/Award Numbers: PP0011‐114754,

PP00P1‐135143; Independent Research

Fund Denmark, Social Sciences Section,

Case No. 275-06-0293

Abstract

This paper studies how competition and vertical structure jointly determine

generating capacities, retail prices, and welfare in the electricity industry.

Analyzing a model in which demand is uncertain and retailers must commit to

retail prices before they buy electricity in the wholesale market, we show that

welfare is highest if competition in generation and retailing is combined with

vertical separation. Vertically integrated generators choose excessively high

retail prices and capacities to avoid rent extraction in the wholesale market

when their retail demand exceeds their capacity. Vertical separation eliminates

the risk of rent extraction and yields lower retail prices.

KEYWORDS

competition, electricity, generating capacities, monopoly, vertical integration

JEL CLASSIFICATION

D42, D43, D44, L11, L12, L13

1

|

INTRODUCTION

Electricity markets around the world have recently been reformed in an effort to improve their performance. In several

countries, legislators have introduced competition into statutory monopoly and imposed mandatory unbundling on

vertically integrated electricity generators.

1

While introducing competition has hardly been controversial, it is fair to say

that the role of vertical structure in the electricity industry is not very well understood. Why should electricity

generators be vertically separated from other layers of the industry, as suggested by legislators (see, e.g., European

Commission, 2010)? Does mandatory vertical separation undermine investments in generating capacity when

generators compete (Joskow, 2006)?

In this paper, we employ a simple model of the electricity industry to study how competition and vertical structure

jointly determine generating capacities, retail prices, and welfare. We consider three vertical layers—generation,

wholesaling, and retailing—and examine four industry configurations that vary with respect to vertical structure and

competition: social optimum, integrated monopoly, integrated duopoly, and separated duopoly. Throughout, we

assume that retail demand is rationed if total retail demand exceeds aggregate capacity. Excess demand thus leads to

“brownouts”(rather than “blackouts,”which would reflect a complete market breakdown) in the event of insufficient

capacity.

Our key result is that—apart from the social optimum—welfare is highest (lowest) if competition is combined with

vertical separation (vertical integration, respectively). The integrated monopoly yields an intermediate level of welfare.

Our analysis, thus, supports the view that introducing competition into the electricity industry should be combined

with vertical separation, even though investments in generating capacity will indeed be smaller than with vertical

integration.

The driving force behind this result is the risk of rent extraction that is associated with vertical integration: An

integrated generator whose capacity is insufficient to serve own retail demand at predetermined prices must buy

electricity in the wholesale market.

2

This drives up the wholesale price to a level at which the troubled generator must

give up its rent. To avoid such rent extraction in the wholesale market, competing integrated generators not only make

large‐capacity investments, but also set excessively high retail prices, thereby avoiding commitments to large retail

sales. The combination of high capacities and low retail demand gives rise to inefficient capacity utilization. Vertical

separation eliminates the risk of rent extraction because vertically separated generators do not need to serve an

uncertain retail demand at retail prices that they have committed to. As a result, both capacity investments and prices

are lower, and welfare is higher, even though consumers are rationed in high‐demand states when demand exceeds

capacity. Intuitively, the result follows because, from a welfare perspective, it is better to reduce demand in high‐

demand states only (by rationing) than to reduce demand in all states (by higher prices).

The risk of rent extraction emerges from the interplay of three ingredients of our analysis: (a) uncertain retail

demand; (b) retailers must commit to retail prices before the wholesale price is determined; and (c) the wholesale price

increases when demand exceeds capacity. In our view, these are natural ingredients of any model of the electricity

industry that studies multiple vertical layers. It should be clear, though, that the risk of rent extraction is mitigated

when there are more than two generators, in which case competition among generators may prevent the wholesale

price from increasing even if one of the firms lacks sufficient capacity. However, if all generators must offer some of

their capacity to satisfy aggregate demand, the risk of rent extraction persists.

Our paper contributes to two related strands of literature. First, we add to the extensive literature on the impact of

demand uncertainty on capacity choices (Boom, 2002; Borenstein & Holland, 2005; Castro‐Rodriguez, Marín, & Siotis,

2009; Drèze & Sheshinski, 1976; Gabszewicz & Poddar, 1997; Grimm & Zoettl, 2013; Murphy & Smeers, 2005; von der

Fehr & Harbord, 1997; as well as Fabra, von der Fehr, & de Frutos, 2011). The key difference to these papers is that we

study the role of vertical structure for capacity choices when demand is uncertain.

Second, we add to the literature on the role of vertical structure in the electricity industry, which has either focused

on the loss of economies of scope from (mandatory) vertical separation (see, e.g., Kwoka, 2002; Kwoka, Pollitt, &

Sergici, 2010), or interpreted vertical integration into the retail market as forward contracts and analyzed their effects

on wholesale prices (see, e.g., Bosco, Parisio, & Pelagatti, 2012; Bushnell, 2007; Bushnell, Mansur, & Saravia, 2008; de

Frutos & Fabra, 2012; Mansur, 2007).

3

None of these papers studies the role of vertical structure in determining

generating capacities. Baldursson and von der Fehr (2007) analyze the effect of vertical integration on the performance

of long‐term and spot markets when spot market prices are uncertain and both independent retailers and electricity

generators are risk averse. They find that vertical integration impairs market performance by increasing the gap

between contract prices and expected spot prices. This effect disappears if agents are risk neutral as in our framework.

The paper closest to ours is Boom (2009), which uses a similar analytical framework. There are two important differences

to this paper. First, we presume that excess demand leads to rationing and brownouts (rather than blackouts and complete

market breakdown). In doing so, we eliminate extreme punishments from blackouts as the driving force behind investment

and pricing decisions. Second, we introduce the separated duopoly configuration, which is indispensable for studying how

competition and vertical structure jointly determine generating capacities, retail prices, and welfare.

The remainder of the paper is structured as follows. Section 2 introduces the model, and Section 3 characterizes the

equilibrium outcomes in the various market configurations. Section 4 provides a ranking of the market configurations

in terms of capacities, retail prices, and welfare, and discusses the role of rent extraction in determining this ranking.

Finally, Section 5 offers conclusions and directions for future research.

2

|

THE MODEL

2.1

|

Vertical structure

We consider a stripped‐down model of the electricity industry with three vertical layers: generation, the wholesale

market, and the retail market (see Figure 1). There is ex ante uncertainty about the level of retail demand, which is

represented by the demand shock ≥ε0. Before the demand uncertainty is resolved, generators

A

and

B

must choose

capacities

k

A

and

k

B

, and retailers Cand

D

must commit to retail prices r

C

and r

D

at which they are willing to serve

demand. After the demand uncertainty is resolved and retailers have learned their demands

d

rrε(, ,

)

CC D

and

d

rrε(, ,)

DC D , retailers buy electricity from generators on the wholesale market. An auction determines the wholesale

price

pp p(,

)

AB

and the quantities

≤yp p k(, )

iAB i

,

i

A

B

=,

, which generators supply to the grid, where

pp,

AB

denote the

BOOM AND BUEHLER

|

211