Many countries used to allow firms to engage in anticompetitive practices as long as they registered their agreements with a government authority. This was the case in several European countries, such as Denmark, Finland, Norway, and Sweden after World War II; or the United States under the National Industrial Recovery Act (NIRA).
In Austria, cartels were legal until the country’s EU accession in 1995. This paper. available here, examines archival material on various types of registered horizontal cartels in Austria to learn about their inner working. It undertakes a content analysis of these legally binding cartel contracts with a view to identifying different collusion methods.
In short, the authors find that these cartel agreements addresses those issues that the academic literature has identified as potential obstacles to sustaining collusion over time. In particular, the agreements set up compensation schemes, reporting requirements, rules for entry and exit, and mechanisms to ensure quick and credible punishment of cartel deviation. The paper is structured as follows:
Section 2 describes how cartel registration worked in Austria.
Under the 1951 cartel law, as updated in 1972, cartel agreements were not prohibited in Austria – instead, they were merely required to be registered with a court. To understand this, one must remember that, after World War II, cartels were considered a stabilising force in the economy. Austria’s market regulation was based on a system of ‘‘social partnership’ under which economic agents were represented in central organisations. Membership of the Chamber of Labour and the Chamber of Commerce was mandatory, and these chambers represented the interests of its members. In addition, employees were represented by the Trade Union Federation, and farmers by the Chamber of Agriculture. These four major interest groups were members of the ‘‘Parity Commission’’, an informal body that sought to arrive at compromises on price and wage issues. These chambers had the power to intervene in the cartel registration procedure – as well as in broader efforts to cap the prices of a number of products.
To legalise a cartel, cartelists had to write a contract defining their cartel arrangement and outlining its economic justification (even though this latter requirement was never used to refuse registering a cartel). The contract then had to be submitted before a three-judge panel at the cartel court. Upon approval, the contract was filed with the cartel register, upon which it became a legally binding agreement. Individual cartels could only be prohibited when the contract on which they were based was deemed immoral under the Austrian civil code – a concept that ensured that bid rigging and restrictions on cartel exit were prohibited.
At the same time, while unregistered cartels could be sanctioned, there were only two prosecutions from 1951 to 1995.
Section 3 describes the archived cartel registry and the sample used by the authors in their analysis. The registry lists c. 125 cartel folders, comprising 149 agreements. After consolidating consecutive (i.e. agreements that extend the duration of a cartel contract) and add-on agreements (i.e. agreements that added new members to an existing cartel contract), and removing contracts that amounted to mergers, the authors were left with 99 contracts, of which 80 refer to purely horizontal arrangements.
Section 4 describes the main cartel agreement clauses identified in the sample. This section comprises the bulk of this paper.
It begins by identifying four main types of registered horizontal cartels – price-fixing, quotas, market sharing, and payment conditions cartels.
– 41 cartels (51.3% of the total) contain price-fixing clauses, but only seven were pure price-fixing cartels. Since a cartelist can easily undermine high prices by granting favourable payment conditions, it is unsurprising that 27 of the 41 price-fixing cartels also included payment conditions clauses.
– 46.3% of the cartel contracts contain quota clauses, even if two-thirds of these arrangements also adopt other collusion methods. Quota cartels are prevalent in the manufacturing sector, and the high percentage of quota cartels reflects the preponderance of the manufacturing sector in the sample of registered cartels.
– Few (26) cartel contracts contain a market sharing clause, and even fewer pure are pure market sharing cartels, perhaps because pure customer allocation cartels were unlawful.
The paper also seeks to identify the orientation and size of collusion. 78 of the 80 registered cartels were seller cartels. The only two cartels that are not at all sales oriented were a pure quota, and a quota/specialisation cartel. At the same time, cartels often operated on multiple directions. For example, roughly a third of pure quota cartels were also buyer cartels; and more than one fourth of the cartels involving market sharing were export-oriented. As regards size, the number of participants in quota cartels is significantly smaller in terms of both average and maximum number of participants.
Regarding cartel governance, while there were six cartels without a formal decision-making body, they all had four members or less, and five were market-sharing cartels. The most common decision-making body in cartel arrangements was a plenary meeting, with some decisions delegated to a committee, the executive officer or the authorised representative—usually a lawyer from a small set of Austrian law offices dealing with cartel issues. In slightly more than one half of sampled cartels, day-to-day management was handled by a staffed office; 35% of cartels relied on independent auditors for investigations; and a joint sales company was established to maintain control over the cartel arrangements in about 15% of cartels. Only in a quarter of cartels was the plenary meeting responsible for fixing prices or setting output. However, one has to keep in mind that coordination on prices was feasible in Austria via the subcommittee on prices or via adherence to external reference prices. In almost a third of all cartels, the plenary meeting was in charge of approving entry and exclusion from the cartel.
Mechanisms to sanction deviations are necessary to sustain cartel stability. The authors document three forms of punishing deviators: a formal warning without any fines, monetary penalties, and exclusion from the cartel. Of all 80 cartel contracts, only 59 explicitly address the issue of punishment; however, one should remember that cartel members could be taken to court for violating the cartel contract. Additionally, it was also common for an arbitration panel to be entrusted with addressing internal disputes between the cartelists.
Section 5 summarises the key findings by collusion method.
This section contains a rather detailed discussion of the characteristics of each type of cartel, which is too long for me to reproduce here. I will therefore limit myself to some insights I found interesting.
Registered cartels included only a small subset of firms active in the Austrian economy. 64 cartels, and thus 80.8% of the sample, were active in manufacturing, ten cartels were in the trade sector, and only six cartels operated in the services sector. More than half of the cartels used more than one method of collusion. For instance, roughly one third of all cartels combined price and payment condition clauses. Market sharing agreements by region, product or customer stand out for the simplicity of the agreements’ terms. Price fixing cartels were comparatively large and often employed other methods of collusion to prevent competition along dimensions other than price. Payment condition cartels primarily tend to have the most complex agreements, as they rely on information being provided upon request and explicit punishment schemes to prevent secret price cuts – which may explain why they were relatively rate.
This paper provides a nice analysis of how some of the main types of cartels – with the exception of bid rigging arrangements – operate in practice. One limitation of the study that the authors correctly identify is that comparing cartels in legal and illegal environments is not straightforward. Legal cartels could openly engage in practices such as compensation, punishment, and information transfers, which illegal cartels must hide. Legal cartels can openly instruct all employees to behave according to the agreement, whereas in illegal cartels it is less clear who is informed and bound by the agreement. Nonetheless, this paper provides some valuable insights into how cartels operate in practice, and is part of a burgeoning and very interesting literature on the topic.