#86 - Agreements, Collusion And Parallel Conduct - Competition Law

Furse on agreements, collusion and parallel conduct

• Horizontal agreements may raise concern that competition is being harmed, but may be difficult to detect

• Problems of policing horizontal conduct may be exacerbated in oligopoly markets where there are a few competitors

• Cartel agreements may be difficult to put in place, but they have long-term success

• Vertical agreements are less likely to raise competitive concern, unless they are linked to the exercise of market power, or contribute to the exclusion of competitors from a market

Horizontal restraints

• Horizontal agreements are those between firms at the same level of production or distribution

• In US, antitrust rules that price fixing is condemned per se, and is not subject to a rule of reason Trans-Missouri Freight Assoc

• The more concentrated a mkt is, the more likely that firms will be able to successfully dampen competition via collusion

  • Note suggestion by Cournot in 1838 that even assuming independent decisions made by oligopolists, prices in such a market would be higher than in perfectly competitive ones

  • They achieve a “Nash-Cournot” equilibrium

    Cartels

• Cartel = “Explicit arrangement designed to eliminate competition”

• Perfect cartel would be one in which the group as a whole set production where MC for the group equalled MR

  • I.e. the cartel would behave as a single-firm monopoly

• Two major problems face any cartel:

  • Agreement

  • Adherence

    Agreement

• Several factors make it difficult for members to reach this

• Product differentiation may mean members have to agree on a complex pricing scheme rather than a single price

  • Note this is a problem for OPEC, as oil is actually produced in different grades

• Larger firms, which can benefit from economies of scale, may want lower prices than smaller firms

  Adherence

• If a cartel is successful in restricting its joint output and raising price, it creates an incentive for individual member firms to cheat, expand their outputs and undermine the cartel

  • A single firm will always profit by cheating on the cartel

  • I.e. cartels are inherently unstable

• Cf Game theory: prisoner’s dilemma – a dominant strategy leads to a sub-optimal outcome

• Only if the two prisoners can exchange information during the process, and are prepared to forgo some short-term benefit in order to improve their collective position, can it be assumed that the outcome will be optimal to the two

  Action by authorities

• Given the difficulty of detecting cartels, most authorities attempt to implement measures designed to destabilise them

  • Largely based on pioneering work of George Stigler in 1964

    • Focused on fact that much cartel activity is secretive

    • Firms will be faced with a reduced demand as prices rise

    • Difficult for firms to determine whether demand reductions are the result of this or of cheating by other members

  • One of the tasks of comp policy is to make info exchange more risky – availability of hard physical evidence of cartelisation normally brings swift condemnation

• Activities of trade associations tend to be closely scrutinised by comp authorities

• Some argued that where cartels exist, they may be limited in harm

  • But recent evidence suggests otherwise, i.e. 2006 study showing overcharge may be as high as 40%

    Price leadership

• It can be difficult to determine whether prices are being maintained across a cartel or whether price leadership is present

• Markham – there are 3 categories of price leadership:

  1. Dominant firm leadership – it is likely to be able to set prices as if it were a monopolist, and small firms have little to gain from diverging much

  2. Barometric price leadership – characteristic of mkts where price leader changes frequently, response to any change in price tends to be less swift than in the situation where there is a dom firm, cf Zinc Powder Group

  3. Markets where the product is homogenous and there are few producers facing similar costd – poses greatest problem for comp authorities; see UK in petrol industry

     Vertical restraints

• In case of most goods, there is a chain of production before the product reaches the customer

  • Vertical agreement is one between firms at different stages of production

• A vertical agreement is to some extent a substitute for vertical integration

  • Therefore surprising that EC, early on, attacked an agreement between a manufacturer and a distributor as being anti-competetive

    • Consten and Grundig, 1966

• Chicago school argues broadly that all vertical restraints should be lawful

  • But today, economists are keen not to generalise

• Argument in favour of examining vertical restraints is that while they may encourage inter-brand competition, they may restrict intra brand competition (i.e. between two sellers of nike shoes)

  • This may not be a significant problem if the buyer has wide choice

    OFT Report puts forward 3 considerations

1. Is there horizontal mkt power either at the level of the manufacturer or the level of retailer?

2. Is consumer likely to be significantly affected by the restriction?

3. Is the result of the restriction to generate efficiency gains?

• It is commonly assumed that vertical restraints are necessarily introduced to benefit the party higher up the chain of production but this is not always the case

  • One of the strongest forms of vertical restraint is Retail Price Maintenance (RPM)

    • Resellers are restricted in their ability to set prices

  • Usually the manufacturer or supplier specifies a min price below which the reseller may not sell

  • Immediate beneficiary tends to be the retailer

    • This was the only act condemned per se in the UK by the 1998 comp act

      Art 81 EC

Summary

• 81 applies to coordinated conduct which appreciably restricts competition]

• Applies equally to agreements, decisions by associations of undertakings and concerted practices

• Most important Q is whether there is in the conduct a prevention, restriction, or distortion of comp within art 81(1)

• Some forms of conduct, such as horizontal price fixing, are per se prohibited

• In others, such as vertical dist. Agreement, an analysis must be conducted into the competitive effect

• There must be an effect on trade between MS

• 81 has direct effect

• There is an exception under 81(3) where the restrictive conduct leads to certain benefits

• Breach renders void any offending agreement, and can lead to penalties, damages and injunctions

  Intro

• Art 81 is intended to apply to any agreed coordinated conduct, however structured

  • More concerned with the economic impact than its legal form La technique Miniere

• The phrase “agreements between undertakings....” should be interpreted widely, to encompass any multilateral conduct

  • Note 81(1) is sufficiently precise as to be directly effective

  • But under 81(3) cases can be exempted on an individual basis, or en bloc

    • This is similar, but not the same as the “rule of reason” used in US in relation to section 1 Sherman act

      81(1)

• Any conduct falling within its ambit will be illegal

  • This can be relied upon by injured competitors

• Commission does not need to distinguish between forms of breach

  • Has even refused to do so in Cartonboard “it may not even be feasible or necessary to make any such distinction [between agreement or concerted practice]”

    Agreement between undertakings

• No requirement that a formal contract be in place

  • But the article cannot at any time be applied to entirely unilateral conduct, cf Bayer

• Polypropylene: an agreement exists “wherever there is the necessary consensus between parties determining the lines of mutual action or abstention from action in the market”

  • Apparently unilateral conduct can lead to 81 infringement, as in Johnson and Johnson – where dealers knew what the position was, and hence prohibitions formed an “integral part of agreements” within the meaning of 81(1)

    Decisions by associations of undertakings

• Standard practice for those in an industry to belong to an assoc that acts on behalf of its members

  • I.e. industry wide promo campaigns, public education, mkt research, standard setting etc

• These may also lead to collusion though

  • I.e. Roofing Felt 1986 – agreement between 7 Belgian asphalt producers to adopt a price list, and set minimum prices for all roofing

• Note that where an assoc is found to exist, the steps that it takes do not need to be binding on all members

  • See Verband

  • “regardless of what its precise legal status may be [the recommendation] constituted the...applicant’s resolve to coordinate the conduct of its members”

  • See also Fenex, where members were “urgently recommended” to adopt a tariff increase of 5%

• Sometimes, the rules themselves of a trade assoc will be such as to make its creation or membership a breach of art 81(1), i.e. in National Sulphuric Acid Assoc

  Concerted Practices

• This is the most nebulous of the categories and covers a wide range of conduct

• First significant EC case was Dyestuffs

  • ECJ upheld a commission decision where undertakings were condemned on basis of collusion

• This was upheld in Suiker Unie, where court set the test as being

  • “concerted practice refers to a form of coordination between undertakings which, without having been taken to the stage where an agreement properly so-called has been concluded, knowingly substitutes for the risks of competition cooperation in practice between them which leads to conditions of comp which do not correspond to the normal conditions on the mkt”

• Difficult to detect, particularly where market is oligopolistic

• It is the method, rather than the result which is being condemned

  • But it is difficult to distinguish between...