Rent sharing to control noncartel supply in the German cement market

• DOI: http://doi.org/10.1111/jems.12234
• Published date: 01 March 2018
• Author: Joseph E. Harrington,Kai Hüschelrath,Ulrich Laitenberger
• Date: 01 March 2018

Received: 17 October 2016 Revised: 29 July 2017 Accepted: 10 October 2017

DOI: 10.1111/jems.12234

ORIGINAL ARTICLE

Rent sharing to control noncartel supply in the German

cement market

Joseph E. Harrington Jr.1Kai Hüschelrath2,4 Ulrich Laitenberger3,4

1Department of Business Economics & Pub-

lic Policy,The Wharton School, University

of Pennsylvania,Philadelphia, PA, USA

(Email: harrij@wharton.upenn.edu)

2Universityof Applied Sciences, Faculty

of Business and Economics, Blechham-

mer 9, D-98574, Schmalkalden, Germany

(Email: hueschel@hs-sm.de)

3Télécom ParisTech,Dépar tement Sci-

ences Sociales et Économiques, 46 Rue

Barrault, 75013, Paris, France (Email: laiten-

berger@enst.fr)

4ZEW Centre for EuropeanEconomic

Research, Mannheim, Germany

Abstract

A challenge for many cartels is avoiding a destabilizing increase in noncartel sup-

ply in response to having raised price. In the case of the German cement cartel that

operated over 1991–2002, the primary source of noncartel supply was imports from

Eastern European cement manufacturers. Testimonies in a private enforcement case

have claimed that the cartel sought to control imports by sharing rents with interme-

diaries in order to discourage them from sourcing foreign supply. Specifically, cartel

members would allow an intermediary to issue the invoice fora transaction and charge

a fee even though the output went directly from the cartel member's plant to the cus-

tomer. We investigate this claim by first developing a theory of collusive pricing that

takes account of the option of bribing intermediaries. The theory predicts that the

cement cartel members are more likely to share rents with an intermediary when the

nearest Eastern European plant is closer and there is more Eastern European capac-

ity outside of the control of the cartel. Estimating a logit model that predicts when

a cartel member sells through an intermediary, the empirical analysis supports both

predictions.

1INTRODUCTION

Collusion is not easy. The prospectof reaping large profits from raising the market price can be jeopardized by several sources of

instability.First, there is inter nal instability coming from a member violating the collusive agreement. This act of noncompliance

is typically for the purpose of gaining a larger share of the market than had been allocated to it. Second, there is external

instability in the form of alternative sources of supply. When the cartel raises price and the cartel is not all-inclusive, those

firms outside of the cartel will often seek to increase their sales by undercutting the collusive price. And, even when all existing

suppliers are members of the cartel, a higher market price can attract entry. Third, a cartel must avoid detection by customers

and the competition authority which would bring an end to collusion. In sum, effective collusion requires that cartel members

are sufficiently compliant with the collusive agreement, noncartel members do not significantly expand supply, and the cartel

avoids detection.1

The project was financially supported by the State Government of Baden-Württemberg,Ger many, through the research program Strengthening Efficiency and

Competitiveness in the European KnowledgeEconomies (SEEK). We are grateful to Cartel Damage Claims (CDC), Brussels, for providing us with the data set

and to the coeditor, two anonymous referees, Sven Heim, PatrickLeg ros, Christine Zulehner,and par ticipants at the 12th CRESSE International Conference

on Competition & Regulation for helpful and constructive comments. We also thank Cung Truong Hoang forexcellent research assistance. Hüschelrath was

involvedin a study of car tel damageestimations which was financially supported by CDC. The study is published in German (K. Hüschelrath, N. Leheyda, K.

Müller, and T.Veith (2012), Schadensermittlung und Schadensersatz bei Hardcore-Kartellen: Ökonomische Methoden und rechtlicher Rahmen, Baden-Baden).

The present paper is the result of a separate research project.

J Econ Manage Strat. 2018;27:149–166. © 2017 WileyPeriodicals, Inc. 149wileyonlinelibrary.com/journal/jems

150 JOURNAL OF ECONOMICS & MANAGEMENTSTRATEGY

Let us consider the second source of cartel instability: increased supply by firms that reside outside of the cartel. There

are numerous episodes for which an expansion in noncartel supply either greatly impacted the profitability of collusion or

even resulted in the demise of the cartel. A notable example is the vitamin C cartel of the early 1990s.2Formed in 1991, it

comprised the four largest producers who in aggregate had 87% of global sales. Of particular relevance, the cartel excluded

Chinese manufacturers who had a market share of 8% at the time. The cartel implemented a 30% increase in prices from 1990

to late 1993, in response to which it lost 29% of global sales to Chinese suppliers, who tripled their sales, and other fringe

producers. With the erosion of the cartel's share of the global market, prices subsequently fell by 33% from the end of 1993 to

1995. The cartel's last formal meeting occurred in August 1995. The failure to control the growth of noncartel supply resulted

in cartel death.

Cartels are well aware of the threat of noncartel supply and they have deployed four general methods for handling it, which

we refer to as takeover, starvation, coercion, and bribery. The takeover approach is the most straightforward and probably the

most effective (though not necessarily the most profitable). Here, the cartel takes control of the sources of noncartel supply

by acquiring noncartel suppliers or the assets necessary to produce. A starvation strategy curtails noncartel supply by taking

control of an essential input or technology. Coercion refers to aggressive practices—such as a targeted price war—that harms

noncartel suppliers with the intent of either inducing them to constrain their supply or join the cartel or exit the market. Rather

than use the stick, a bribery strategy uses the carrot by sharing collusive rents with noncartel suppliers if they agree to limit their

expansion of supply.As t he identification of these fourstrategies is, to our knowledge, new to the literature, Section 2 illustrates

them with a collection of cartel cases.3

The focus of this paper is exploring how noncartel supply was handled in the case of the German cement cartel which lasted

from 1991 to 2002. The primary threat was imports from cement manufacturers in Eastern Europe, specifically, plants located

in the Czech Republic, Poland, and Slovakia. For legal and logistical reasons, importation from those countries into Germany

required the use of German companies acting as intermediaries. The claim was made in the context of a private litigation case4

that, in order to avoid intermediaries bringing the cement of Eastern European suppliers into the German market, the German

cement cartel shared some of the collusive rents with those intermediaries. In light of the four methods for handling noncartel

supply,this conjectured strategy could be cast as either starvation—by controlling a keyinput to foreign cement manufacturers—

or bribery—in that intermediaries were bought off.

For the purpose of examining the validity of the claim that the German cement cartel bribed intermediaries in order to

limit noncartel supply, this paper develops a theory of collusive behavior that encompasses the option of sharing rents with

intermediaries. The theory is then taken to data and we find evidence in support of the theory's predictions. Hence, we find

empirical support for the claim that the German cement cartel bought off intermediar ies forthe pur pose of limiting imports and

thereby constraining sources of noncartel supply.

The paper is organized as follows. Section 2 reviews how various cartels have controlled noncartel supply. Section 3

provides a general description of the German cement cartel including an overview of primary threats to its stability with

a focus on the role of intermediaries. In Section 4, a theory of collusive pricing in the presence of intermediaries and

noncartel suppliers is developed which allows for the option of sharing rents with an intermediary in exchange for them

not sourcing noncartel supply. That theory generates a set of hypotheses that are then tested in Section 5. Section 6

concludes.

2CARTEL STRATEGIES FOR CONTROLLING NONCARTEL SUPPLY

Suppose a cartel increases price in the presence of some suppliers who are not part of the cartel. The likely response of those

noncartel suppliers is to undercut the cartel's price and expand supply which has the potential for substantively undermining

the profitability of collusion. In responding to the expansion of noncartel supply, cartels have pursued four strategies: takeover,

starvation, coercion, and bribery. Examples of each of these are provided below.

Takeover: A cartel curtails noncartel supply by acquiring the noncartel suppliers or the assets used to provide that supply.

The five global producers of aluminum formed a cartel in 1900–1901 which lasted until 1908 when a recession and entry

caused the cartel's collapse. The established firms then went about acquiring nine recent entrants after which they reestablished

the cartel in 1912.5

Members of the international steel cartel in the 1930s acquired fringe firms at prices based on “their nuisance value to the

cartel” which apparently exceeded a valuation based on their projected earnings.6

The electrical and mechanical carbon and graphite products cartel operated over 1988–1999 and struggled with noncartel

suppliers known as "cutters" which would purchase carbon blocks from the cartel members and then produce final products