1. London 22 Nov 2005 Modernization of Article 82 Lars-Hendrik Röller * Chief Competition Economist European Commission CLA and BIICL Conference on Article 82 London 22 November 2005 *The views expressed are those of the author and do not necessarily reflect those of DG COMP or the European Commission

2. London 22 Nov 2005 outline Why the review? Why the review? The EAGCP report The EAGCP report effect-based approach effect-based approach rule-of reason vs. per se approach rule-of reason vs. per se approach role of dominance role of dominance Current review – overview Current review – overview Other issues Other issues

3. London 22 Nov 2005 Why the review of Article 82 current approach too restrictive current approach too restrictive In the past: form-based approach and per se rules (loyalty rebates in Michelin II) In the past: form-based approach and per se rules (loyalty rebates in Michelin II) need to limit false positives that chills pro-competitive conduct need to limit false positives that chills pro-competitive conduct need to limit false negatives that allows conduct that lead to consumer harm need to limit false negatives that allows conduct that lead to consumer harm effect-based analysis effect-based analysis in order to better discriminate between harmful and pro-competitive conduct (limit false positives/false negatives) in order to better discriminate between harmful and pro-competitive conduct (limit false positives/false negatives) to identify relevant theory of harm & the relevant empirical evidence to identify relevant theory of harm & the relevant empirical evidence should provide legal certainty and consistency should provide legal certainty and consistency may require significant investigation (time and resources) may require significant investigation (time and resources) clarifies and creates consistency clarifies and creates consistency approach consistent with Art. 81, merger control, and state aid action plan approach consistent with Art. 81, merger control, and state aid action plan

4. London 22 Nov 2005 The EAGCP report not form-based, but effect-based not form-based, but effect-based categories of conduct (predation, discrimination, rebates, tying, refusal to deal) can have the same effect categories of conduct (predation, discrimination, rebates, tying, refusal to deal) can have the same effect effect of foreclosure (same, horizontal, vertical markets) on consumers (as a proxy for competition -multiple, present and future) effect of foreclosure (same, horizontal, vertical markets) on consumers (as a proxy for competition -multiple, present and future) “the story” = role of robust theory and facts “the story” = role of robust theory and facts more rule-of-reason, less per-se rules more rule-of-reason, less per-se rules rule-of-reason better done under an effect-based approach rule-of-reason better done under an effect-based approach no need for separate dominance under an effect-based approach no need for separate dominance under an effect-based approach dominance and its abuse => interrelated dominance and its abuse => interrelated case law vs. legal norm of Art. 82 => abuses of dominance case law vs. legal norm of Art. 82 => abuses of dominance

5. London 22 Nov 2005 Current review of Article 82 (some key elements) abuse-specific approach, but effect based abuse-specific approach, but effect based competition not competitors, consumer welfare competition not competitors, consumer welfare all effects (likely & actual, short & long-term) all effects (likely & actual, short & long-term) as efficient competitor test as efficient competitor test on dominance on dominance necessary condition necessary condition more economics in dominance assessment more economics in dominance assessment => role of the case law and the ECN

6. London 22 Nov 2005 rule-of-reason vs. per-se rules need presumptions & safe harbors as a screen need presumptions & safe harbors as a screen green light zones & red light zones green light zones & red light zones “rule of reason” analysis for some cases (follow EAGCP approach) “rule of reason” analysis for some cases (follow EAGCP approach) predictability vs. precision: is there a trade-off ? predictability vs. precision: is there a trade-off ? Vickers argument Vickers argument ultimate trade-off between precision and predictability (and enforcement costs) ultimate trade-off between precision and predictability (and enforcement costs) predictability of rules predictability of rules

7. London 22 Nov 2005 Is there a trade-off in Article 82? Predictability Precision rules Rule-of-reason

8. London 22 Nov 2005 Overall effect-based approach effect-based approach consumer orientation consumer orientation pre se rules and rule of reason pre se rules and rule of reason dominance necessary condition dominance necessary condition some other issues……..

9. London 22 Nov 2005 the role of dominance only dominant firms are subject to Article 82 only dominant firms are subject to Article 82 advantage: bright lines + legal certainty + enforcement costs advantage: bright lines + legal certainty + enforcement costs disadvantages: under-enforcement disadvantages: under-enforcement abuse of pre-existing dominance abuse of pre-existing dominance But market power may create significant anticompetitive conduct => acquiring dominance through such conduct might be abused But market power may create significant anticompetitive conduct => acquiring dominance through such conduct might be abused how much economic analysis under dominance, given that effect-based analysis is to follow? how much economic analysis under dominance, given that effect-based analysis is to follow? how do we determine that a firm has substantial market power? how do we determine that a firm has substantial market power? use of market share? use of market share? analysis of barriers to entry? analysis of barriers to entry? price-cost margin? price-cost margin? integrated approach? integrated approach? what is the correct screen? what is the correct screen? can a non-leader be a dominant firm? can a non-leader be a dominant firm? should the path to dominance matter in the competitive assessment? should the path to dominance matter in the competitive assessment?

10. London 22 Nov 2005 an over-arching test  no reason to believe that one single test gives best answer in all cases  as-efficient competitor test  cost data  potentially efficient  competitive constraint, even if inefficient  non-price competition  product differentiation probably better to apply different (structured) rules, including shortcuts, for different practices   but strive for consistent framework   to show: capability to foreclose and market distorting effect

11. London 22 Nov 2005 predation How much form-based is there in the case law? How much form-based is there in the case law? AKZO rule still basic framework AKZO rule still basic framework Below AVC/AAC rebutable presumption of abuse Below AVC/AAC rebutable presumption of abuse Between AVC and ATC need to prove strategy of foreclosure (“intent”) Between AVC and ATC need to prove strategy of foreclosure (“intent”) No separate requirement to prove recoupment beyond dominance No separate requirement to prove recoupment beyond dominance Except perhaps in specific cases (e.g. collective dominance) Except perhaps in specific cases (e.g. collective dominance) Predation: likely foreclosure, sacrifice, recoupment Predation: likely foreclosure, sacrifice, recoupment

12. London 22 Nov 2005 ….. efficiency defense efficiency defense abuses have two effects abuses have two effects burden of proof (burden of production) burden of proof (burden of production) integrated into conduct (pricing efficiencies) integrated into conduct (pricing efficiencies) market definition market definition defining the relevant product or geographic market is a legal requirement defining the relevant product or geographic market is a legal requirement by definition the dominant firm does not face any competitive constraint. So does the dominant firm constitute a market on its own? by definition the dominant firm does not face any competitive constraint. So does the dominant firm constitute a market on its own? applying the SSNIP test would often result in a broader market definition applying the SSNIP test would often result in a broader market definition evaluate directly the impact of the exclusionary abuse should help delineate the boundaries of the relevant market evaluate directly the impact of the exclusionary abuse should help delineate the boundaries of the relevant market exploitative vs. exclusionary abuse exploitative vs. exclusionary abuse is monopoly pricing an abuse of dominant position? is monopoly pricing an abuse of dominant position? in market economy, risk of reducing firms’ incentives to be more efficient or innovative. Merger control, IP rights and regulation of natural monopolies can better limit false positives and negatives in market economy, risk of reducing firms’ incentives to be more efficient or innovative. Merger control, IP rights and regulation of natural monopolies can better limit false positives and negatives should there be guidelines? (how much do we know?) should there be guidelines? (how much do we know?)

13. London 22 Nov 2005 Convergence with the US? institutional differences with EU institutional differences with EU US approach may put more emphasis on green zones US approach may put more emphasis on green zones economics is a common language (same questions, same methodology) economics is a common language (same questions, same methodology) answers to the same case may still be different answers to the same case may still be different

14. London 22 Nov 2005 The role of economics DG-Comp – more emphasis on economic effects DG-Comp – more emphasis on economic effects Judicial Review – recent judgments Judicial Review – recent judgments ECN - consistency ECN - consistency

15. London 22 Nov 2005 conclusion pragmatic approach pragmatic approach important improvements important improvements legal certainty legal certainty consistency consistency competitiveness competitiveness