AMSTERDAM, OCTOBER 2006 MATTHIJS VISSER

EconomicsRBB

Abuse Of Dominance

Vereniging voor Mededingingsrecht

2

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Overview

• The Assessment of Dominance

• Exploitative abuses

• Foreclosure: The Discussion Paper of the Commission

• Specific categories of abuse:

– Rebates

– Predation

3

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

The Assessment of Dominance (1)

Economic definition of dominance: ability to raise price above “competitive level”

But what is the competitive price?• In theory?

Marginal Cost (“perfect competition” model)? Average Cost (no “super-normal” profits)?

• In practice, what’s observable?

If we could identify and observe the competitive price, dominance assessment would be trivial

4

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

The Assessment of Dominance (2)

Example: industry with large fixed costs:

• Marginal cost pricing may not permit cost recovery

• Even with perfectly free entry, a sustainable market structure must allow price to be above marginal cost

• Oligopoly structure sustainable in competitive, free entry environment

• More fragmented structure not sustainable

5

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

The Assessment of Dominance (3)

Competitive constraints:

Existing Competitors• Are market shares a good measure?

Potential Competitors (Entry)• How significant are the barriers to entry?

Countervailing Buyer Power• Is the mechanism credible?

6

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Exploitative Abuses

• Extremely limited number of cases

• Fundamental difficulties with exploitation cases: abuse is defined with reference to competitive price level, which is unknown

• Exploitation cases risk being circular and lend themselves to backward induction: observed price is higher than competitive benchmark, hence someone must be dominant, hence must identify market for which this is true (may lead to one firm markets – and additional problems!)

• Competition Authority risks being some sort of price regulator

7

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

The Discussion Paper of the Commission: Objectives

• Introducing a more effects-based approach

• Install consumer welfare and economic efficiency as the guiding principles

• “With regard to exclusionary abuses the objective of Article 82 is the protection of competition on the market as a means of enhancing consumer welfare and of ensuring an efficient allocation of resources. […] In applying Article 82, the Commission will adopt an approach which is based on the likely effects on the market”

8

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Why an Effects-Based Approach?

• The same conduct can be anti-competitive in certain cases and pro-competitive in others

• Effects-based approach provides better results for consumer welfare than form-based approach

– Exclusionary practices are likely to have efficiency-enhancing properties

– Harm to Consumers vs. Harm to Competitors

– Reduce number of false positives (such as Michelin II) and, to a lesser extent, false negatives

9

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Assessment of Discussion Paper

• Effects-based regime calls for coherent theories of harm that are tested against alternative explanations

• The Commission’s Discussion Paper does not move away (enough) from the old, form-based approach

– Numerous exceptions to the “as efficient” competitor test

– Art 81(3) type of defence, burden of proof upon dominant firm

– Importantly, Commission does not require harm to consumers to be shown/proven

10

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Burden of Proof• Discussion Paper proposes two-stage competitive assessment:

(i) Preliminary Finding on basis of prima facie evidence

(ii) Possible reverse based on more detailed assessment justifying the observed conduct or citing efficiency gains

• Balance on burden of proof depends on the application in practice, but wording of DP hints at an overly restrictive regime

• Successful Efficiency defence needs to prove all of the following

– Observed conduct gives rise to efficiency gains

– Observed conduct is indispensable to reach said efficiency gains

– Efficiency gains benefit consumers

– Competition is not significantly reduced

11

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Threshold for Abuse

Firms can exit the market because of:

• Intense competition on the merits, OR

• Anti-competitive behaviour of a dominant firm

An effects-based policy would seek to distinguish these two cases

Note: Dominant firms should be allowed to compete

12

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

The “As Efficient” Competitor Test

The DP introduces a key test for price abuses (Para 63):

“in general only conduct which would exclude a hypothetical ‘as efficient’ competitor is abusive”

Example:

• Domco has a unit cost = € 5• Rival firm has a unit cost = € 7

• Domco price of € 4 would harm an “as efficient” competitor• Domco price of € 6 would not harm an “as efficient”

competitor

This is a useful test!

13

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Should “Less Efficient” Competitors be Protected?

Should competition authorities protect less efficient competitors? (See Para 67)

Example:• Domco has a unit cost = € 5• Rival firm has a unit cost = € 7

• Chain of events:– Rival firm prices at € 8– Domco prices at € 6– Rival cannot compete and is marginalised (or exits the market)– Domco raises price at € 10 (monopoly price)

• Conclusion (erroneous): inefficient competition is better for consumers (in the long term) than no competition at all

14

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Should “Less Efficient” Competitors be Protected?

Protecting less efficient competitors is impractical and inappropriate:

• Domco would have to know its rivals’ cost in order to set its own price

• Being protected from competition authorities, the rival would have no incentive to become more efficient

• Consumers are clearly worse off in the short term since they are denied a lower price

• Consumers are worse off in the long term if lack of incentives for efficient entry

15

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

• Incremental rebate: once you hit the target, discount applies to those units purchased at or above the target

• Rollback rebate: once you hit the target, the lower price relates to ALL units purchased (also called ‘retroactive’ or ‘back to unit one’)

Conditional Rebates: Definitions

Units purchased

List price

Discounted price

extra rebate with rollback when hit target

target

Unit price

List price

Discounted price

extra rebate with rollback when hit target

16

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Rollback Rebates: Competitive Concern

• The dominant firm’s product is a “must have” product

• The dominant firm has a captive base of customers

• The rebate scheme allows the firm to expand its captive base (the “suction effect”)

• Rival firms are unable to sell their products, and are induced to exit the market or permanently reduce capacity

• The dominant firm raises prices in the long term

17

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Suction Effect: Numerical Example

• Supplier offers rebate of 4% on all sales once sales > 1,000

• Regular price per unit is 100

• Buyer “normally” buys 950 units from this supplier.

• What is the effective price of the next 50 units?

• Price with rebate = 1,000 x 96 = 96,000

• Price without rebate = 950 x 100 = 95,000

• Effective price of incremental units = (96,000 – 95,000) / 50 = 20

The suction effect becomes larger the higher the threshold and the higher the rebate

18

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

The DP on Rollback Rebates

According to the Discussion Paper, a rollback rebate system creates a market distorting foreclosure effect if all of the following conditions are fulfilled:

• The threshold is likely to incentivise a “good part” of Domco’s buyers by penalising them if they switch

• The Required Share (RQS) exceeds the Commercially Viable Share (CVS)

• The rebate system affects “if not most, at least a substantial part of the market”

• There are no clear indications of a lack of foreclosure effect such as aggressive or significant entry / expansion or switching by customers

19

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Pro-competitive Effects• Reward promotional efforts of downstream firms,

when their effort is hard to monitor• demand volatile, so share of needs ‘controls’ for

demand shock

• Lowering input costs for downstream buyers and thereby encouraging them to compete on price:• input price large share of downstream costs• small buyers constrain large buyers when sell on in

downstream market)

• Combining appropriate incentives to invest with a buyer’s desire to stock a second supplier• if exclusivity promotes investment, so should near

exclusivity• buyer may value second supplier (differentiation,

second source of supply)

• Reflect buyer specific efficiencies

20

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Anti-competitive Effects – Initial Screens

Dominant Supplier

Final consumers

Rival supplier

‘loyal’ buyers Other buyers?

1: dominance?2: alternative routes to market to become a viable competitor?

Are ‘loyal’ buyers the gateway to the market?

Is self supply viable? (No entry barriers in downstream market)

Is direct supply viable?

3: how does discount scheme affect incentives of buyers?

What is the reward?

How easy to meet target?

•Natural growth?

•Greater promotional effort?

•Direct substitution of Domco for rival?

21

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Predatory Pricing• The Discussion Paper’ defines predatory pricing as

“The practice where a dominant company lowers its price and thereby deliberately incurs losses or foregoes profits in the short run so as to enable it to eliminate or discipline one or more rivals or to prevent entry by one or more potential rivals thereby hindering the maintenance of the degree of competition still existing in the market or the growth of that competition”

• A number of benchmarks “below which no additional proof may need to be brought by the authority because predation can be presumed” are introduced, including

– AAC (Average Avoidable Costs)

– ATC (Average Total Costs)

– LAIC (Long-run Average Incremental Costs)

22

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Predatory Pricing

• The DP essentially re-states the AKZO principles, with few exceptions• AAC benchmark replaces AVC • Still no requirement to show recoupment But recoupment is essential!• The DP leaves too much scope for finding predation above AAC

The AKZO tests revisited:

– Price above ATC can be predatory in specific cases (e.g. selective price cutting)

– Price above AAC but below Average Total Cost (ATC) is predatory if it is part of a plan to eliminate rivals

– Price below Average Avoidable Cost (AAC) is presumed predatory

Price

ATC

AAC

23

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

Cost Benchmarks: Average Avoidable Costs

• Avoidable Cost: the cost that could have been avoided if the firm had not produced a certain amount of output over a certain period of time

Is price less than Average Avoidable Costs ?

• If revenues do not cover avoidable costs, the firm sells at a cash loss Inconsistent with short-term maximisation of profits

• The AAC measure does not depend on accounting conventions, since it does not require establishing whether costs are variable or fixed

• Avoidable Costs tend to increase when a longer timeframe is considered A pricing strategy is more likely to be seen as exclusionary the longer it is maintained

24

MA

TTH

IJS

VIS

SER

VER

EN

IGIN

G V

OO

R

MED

ED

ING

ING

SR

EC

HT

AM

STER

DA

M,

OC

TO

BER

2

00

6EconomicsRBB

I: www.rbbecon.comE: Matthijs.Visser@rbbecon.com