class3-competition and cartels in europe

SSA - 2015 - 1

International MBA Economic Policy and European Integration Stéphane SAUSSIER [email protected]

Slides are downloadable at: http://www.webssa.net

Cartels in Europe

SSA – 2015 - 2

Warm-Up

SSA – 2015 - 3

You are in a monopoly situation (Cont…)

•  Suppose you are the only seller of airline tickets between two Chinese cities

•  Same cost: $5 per unit

•  You face the following demand curve

Price

Quantities

SSA – 2015 - 4

Quantities, Prices, Costs and Profits

Quantity Sold

Price

Total Revenue

Marginal Revenue

Marginal Cost

Total Cost

Total Profit

0 >20 0 0 1 20 20 20 5 5 15 2 18 36 16 5 10 26 3 16 48 12 5 15 33 4 14 56 8 5 20 36 5 12 60 4 5 25 35 6 10 60 0 5 30 30 7 8 56 -4 5 35 21 8 3 24 -32 5 40 -16

Principle: A monopolist in order to maximize its profits should produce such as its marginal revenue equals its marginal cost

SSA – 2015 - 5


• Comparing Marginal revenue and price •  For a monopolist, marginal revenue is less

than price •  This is not the case in a competition situation

•  Depends of the elasticity of the demand curve • How sensible is demand when prices are

varying?

SSA – 2015 - 6

Monopoly vs. Competition

Price

Quantities

Price Quantity Seller’s Profits

Buyer’s Profits

Total surplus

Monopoly 14 4 36 12 48

Competition 5 7 0 63 63

Competition price

SSA – 2015 - 7

You are in a Competition situation Consumers’ surplus

Price

Quantities

SSA – 2015 - 8


Price

Quantities

Consumers’ surplus

Producers’ surplus

SSA – 2015 - 9


Quantities

Consumers’ surplus

Producers’ surplus

Price

Surplus Loss

SSA – 2015 - 10

NOW LET’S CONSIDER CARTELS

SSA – 2015 - 11

You are in a Cartel situation

•  Suppose you are two sellers of airline tickets between two Chinese cities

•  Same cost: $5 per unit

•  You face the following demand curve

Price

Quantities

SSA – 2015 - 12

LET’S PLAY

• What price would you charge if quotas are enforced ?

• What price would you charge if quotas are not enforced and you are near retirement ?

• What price would you charge if quotas are not enforced and you are not near retirement ?

SSA – 2015 - 13

What do we observe?

•  Quantities under a monopoly situation compared to competition?

•  Quantities under a cartel situation compared to monopoly situation?

•  Stability of cartels when quotas are enforced? •  Stability of cartels when quotas are not enforced? •  Stability of cartels when quotas are not enforced but

supposed to stay for a long period?

SSA – 2015 - 14

Theory and facts

CARTELS

SSA - 2015 - 15

Remember: Market Size Matters!

• European leaders always viewed integration as compensating small size of European nations.

•  Implicit assumption: market size is good for economic performance.

« The countries of Europe are too small to give their peoples the prosperity that is now attainable and therefore necessary. They need wider markets » Jean Monnet 1943

« By its size – the biggest in the world – the single market without frontiers is an invaluable asset to revitalize our businesses and make them more competitive. It is one of the main engines of the European Union » J. Delors 1987

SSA - 2015 - 16

The impact of free trade liberalization

•  Result: •  Bigger, fewer, more efficient firms facing more

effective competition

•  Welfare gains (i.e. price decrease)

“Competitive markets make our companies more innovative, more productive and more cost effective, and at the same time deliver lower prices, better quality, new products and greater choice for our citizens. Competitive markets require a strong competition policy, rigorously enforced, and the EU’s state aid rules are an intrinsic part of this” – Neelie Kroes 2008 -- Vice President of the European Commission

Competition does not emerge naturally?!

SSA - 2015 - 17

The impact of free trade liberalization

•  One immediate question •  As the number of firms falls, isn’t there a tendency

for the remaining firms to collude in order to keep prices high?

•  The answer to those questions is “Yes”.

SSA – 2015 - 18

CLASS objectives

Focus on the Cartel strategy and the European solutions to reduce their frequency

1.  How firms might reduce competition by using cooperative agreements?

2.  What conditions are necessary for cartels to work well? •  i.e. Where the European Commission should look

for cartels? 3.  What the European Commission might do to

deal with this issue?

SSA – 2015 - 19

Competition is a problem for firms! •  Aggressive behaviors to destroy competition

•  Look forward to eliminating competitors i.e. Price wars, forclusion, etc.

•  Look forward to avoid market entry by potential competitors i.e. Entry barriers

•  Cooperative strategy – Collusion/Cartels •  «Colluding is the fact to agree (usually secretly) in order to hurt

somebody else» •  In the study of economics and market competition,

collusion takes place when rival companies cooperate for their mutual benefit.

– Example: OPEC – 13 countries deciding of oil production •  Collusion might take several forms

•  Share of the markets; coverage offer; Non aggression pact; Tacit vs. Explicit

Cooperative Strategies: Cartels/Collusion

SSA – 2015 - 20

OPEC

OPEC countries by exercising control and reducing output from where it would be in a uncontrolled, competitive market, are able to shift the equilibrium from the competitive point (P*, Q*) to the "OPEC" point: (P(O), Q(O)), which gives price that is higher than the competitive market price.

What strategy? What limits?

SSA – 2015 - 21

Collusion/Cartel

•  « People of the same trade seldom meet together, even for merriment and diversion, but the conversation ends in a conspiracy against the public, or in some contrivance to raise prices. It is impossible indeed to prevent such meetings, by any law which either could be executed, or would be consistent with liberty and justice. But if the law cannot hinder people of the same trade from sometimes assembling together, it ought to do nothing to facilitate such assemblies; much less to render them necessary ».

—Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations (1776)

☞ Sherman Act strictly forbid such agreements in the US (1890)

☞ Article 85 of the Treaty of Rome (1957) makes it illegal for companies to enter into agreements that restrict or distort competition with the European Union, that is, by means of price fixing or agreements over market share.

WHY ?

“Cartels are harmful to consumers, businesses and to the economy in general” – European Commission, DG COMP

Cartels and social surplus: an impact similar to monopoly!

•  Graphically:

Demand

MC=AC

Qc

Pc

A

Demand

Qc

Pc

Pm

Qm

LOST

AC=MC

Collusion is a way for firms to behave as if they were in a monopoly situation.

- Reduce quantities - Increase price - Increase and share profits

Cartels and social surplus: an impact similar to monopoly!

SSA – 2015 - 24

Should we care about cartel behaviours?

Is this a credible strategy for firms? •  Cartel: Each firm reduces its production in

order to increase prices. •  But each firm has an interest to cheat

(“free riding”) and to let partners pay the cost of collusion (reduction of production) while benefiting of high prices on the market.

•  How such agreements are to be respected?

•  Furthermore, such agreements are prohibited and thus there is a cost of being detected by antitrust institutions

SSA – 2015 - 25

•  Should we care about cartels? (Frequent and stable?)

•  If yes, where should we look at?

•  What do you think are the main determinants for firms to decide to enter or not in a cartel?

•  How would you determine fines against cartels? •  (What determinants would you propose?)

• Discussion

Costs > Gains ! Fines . Probability to be detected > Gains

! Fines > Gains / Proba

SSA – 2015 - 26

This is not a set of simple questions.

Let’s see what a simple economic analysis can teach us.

SSA – 2015 - 27

Should we care about cartel strategy? An economic analysis

• Static approach and stability of collusive behaviour.

•  Collusive strategy leads partners into a prisoner dilemma (Nash 1950).

– Such strategy should never emerge!

(10, 10) (19, -1)

(-1,19) (0,0)

Company 1

Com

pany

2

C

NC

Cooperative Non Cooperative The case of two firms splitting the market in order to share a monopoly surplus of 20.

Nash Equilibrium

J. Nash 1928 -

SSA – 2015 - 28

•  Repeated interactions and stability of collusive strategy. •  Repeated game.

Millipede dilemma!

…


SSA – 2015 - 29

Should we care about cartel strategy? FACTS!

SSA – 2015 - 30


On April 13 2011, consumer goods giants Procter & Gamble and Unilever were fined €315.2m (£280m) by the EC for breaching competition law in the laundry detergent market in eight countries. According to Joaquim Almunia, the European Union's vice-president for competition policy, the three companies began to co-ordinate their efforts in 2002. They “agreed to protect their respective market shares. They also agreed not to decrease prices when decreasing the size of packages, and afterwards, they agreed to a price increase."

SSA – 2015 - 31


Cartel Case :

SSA – 2015 - 32


•  Cartel Case :

SSA – 2015 - 33

Last case

SSA – 2015 - 34

•  Nevertheless we know that such collusive behaviors are numerous!

•  See Emmanuel Combe -Telos (April 2008 – in French) : •  http://www.telos-eu.com/fr/article/

consommateurs_les_ententes_vous_spolient

•  Sometimes very long living cartels. Example: •  (October 1st, 2008): The European Commission has

imposed a total of € 676 011 400 fines on 9 groups for participating in a cartel for paraffin wax in the European Economic Area (EEA) between 1992 and 2005.


SSA – 2015 - 35

•  Fines imposed by the Commission – Sources: DG Competition (April 2013)


SSA – 2015 - 36

Sources: DG Competition – 2014


SSA – 2015 - 37

•  An infinitely repeated game analysis •  Equivalent to a game in which parties

does not know when it will stop •  Millipede dilemma disappears

•  New rules of the game (Trigger strategy) •  Parties agree to cooperate until one

of them cheats. Then there is no way to recover confidence.

•  Profit to be shared is πM=20

•  If there is no cooperation, then profit obtained by firms is normalized to 0

When should we care about cartel strategy? An economic analysis

Nobel Prize 2005

SSA – 2015 - 38

Gains of cooperation for the firms

Gains of no cooperation for the firms 20 + 0 + 0 + …


SSA – 2015 - 39

But delta is a discount rate depending on: With

r: financial annual interest rate; f: frequency of price adjustment; h: probability to see the structure of the market the same at next period and g: the growth rate of the market


€

δ =1

1+rf

h(1+ g)

SSA – 2015 - 40

A collusion price is more probably a equilibrium when •  The frequency with which firms might adjust their prices is high •  The durability of the industry is high •  The growth rate of the industry is high.

•  Examples? •  What activities should be on the

radar screen ?– Water industry?

When should we care about cartel strategy?

SSA – 2015 - 41

Discussion What about

•  The number of partners (market is concentrated or not)? •  The comparability of partners? •  Information about prices and global demand are available or not?

•  Pro-cycle and anti-cycle price wars

The case of Concrete market in Denmark


0

100

200

300

400

500

600

janv-94 avril Jul Oct mars-95 Jun Nov

Ent1

Ent2

Ent3

Ent4

Source: Introduction to Industrial Organization, Luis Cabral, Chapter 8, 2012 – http://luiscabral.org//economics/books/iio2/

SSA – 2015 - 42

Discussion

What about attractive promotional arguments such as :

☞ “If you find a lower price 50 km around, we pay you back (10 times) the difference!”

☞ “Do not delay you consumption. If prices falls in this shop the next coming 30 days, we pay you back the difference”


SSA – 2015 - 43

WHAT ARE THE SOLUTIONS YOU MIGHT THINK OF TO AVOID CARTELS?

SSA – 2015 - 44

As simple as a phone call !

SSA – 2015 - 45

Institutional factors limiting collusion: The US case

•  The antitrust division of the Department of Justice – USA put in place a leniency program (1978)

•  Leniency at DOJ’s discretion, even if the requirements were met

•  Leniency conditioned on request prior to DOJ beginning its investigation

•  DOJ received only one request for leniency

•  Results : Leniency Program failed to uncover a single international cartel

SSA – 2015 - 46

Institutional factors limiting collusion

•  “These three major cornerstones – severe sanctions, heightened fear of detection, and transparency in enforcement policies – are the indispensable components of every effective leniency program.”

•  Scott D. Hammond, Director,,U.S. DOJ Antitrust Criminal Division (2004)

SSA – 2015 - 47

Institutional factors limiting collusion • The antitrust division of the Department of Justice – USA put in place a Leniency program (1978) •  Modified in 1993. New regime

stipulates that the Leniency program might applied if

•  The cartel is denounced before the authority knows its existence

•  Only one restriction: no amnesty for the party who is at the origin of the cartel

•  Results •  In 1999, the amount of fines is higher than

the amount of all fines decided since the Sherman Act!

SSA – 2015 - 48

Penalties and Leniency program in EU •  The leniency policy offers companies

involved in a cartel - which self-report and hand over evidence - either total immunity from fines or a reduction of fines which the Commission would have otherwise imposed on them.

•  It benefits the Commission, allowing it not only to pierce the cloak of secrecy in which cartels operate but also to obtain insider evidence of the cartel infringement.

•  The leniency policy also has a very deterrent effect on cartel formation and it destabilizes the operation of existing cartels as it seeds distrust and suspicion among cartel members.

SSA – 2015 - 49

Penalties and Leniency program in EU •  In order to obtain total immunity under the leniency policy,

•  Be the first one to inform the Commission of an undetected cartel by providing sufficient information to allow the Commission to launch an inspection at the premises of the companies allegedly involved in the cartel.

•  If the Commission is already in possession of enough information to launch an inspection or has already undertaken one, the company must provide evidence that enables the Commission to prove the cartel infringement.

•  The company may not benefit from immunity if it took steps to coerce other undertakings to participate in the cartel.

SSA – 2015 - 50

Penalties and Leniency program in EU •  Companies which do not qualify for immunity

may benefit from a reduction of fines if they provide evidence that represents "significant added value" to that already in the Commission’s possession and have terminated their participation in the cartel.

•  The first company to meet these conditions is granted 30 to 50% reduction, the second 20 to 30% and subsequent companies up to 20%.

•  “First mover effect”

SSA – 2015 - 51

•  The European Commission adopted such a Leniency Program in 1996.

•  In 2001, the amount of fines is higher than the amount of all fines decided since the treaty of Rome (1957) – 10 cartels are condemned - 56 companies – for an amount of 1,836 billion euros

•  France adopted such program in its New Economic Regulation Law (NRE) in 2001.

•  Article L.464-2 in the Trade code stipulates a 50% exemption on fines for companies that do not contest decision and accept to change their future behavior.

•  A complete or partial exemption is also provided for firms denouncing a cartel agreement to the French Competition Council

•  Ex: Christmas Toys case •  Today, many EU members adopted such program

Institutional factors limiting collusion

SSA – 2015 - 52

The Toys Case – 1 – •  Toy manufacturers and toy suppliers

•  The suppliers concerned reached agreements with their distributors in order to fix their products' resale prices in all the retail outlets.

•  During 3 years •  During Christmas time •  These vertical agreements were

accompanied by ‘price monitoring' practices concerning ‘deviant' distributors in order to increase retail prices for ‘problematic toys' and realign retail prices quickly.

•  This was evidenced by numerous errata published by distributors to correct – and systematically increase - prices a posteriori in their Christmas catalogues.

SSA – 2015 - 53

The Toys Case – 2 –

•  Particularly severe for Carrefour: •  « Carrefour, set up a promotional

campaign called ‘Carrefour reimburses the difference times 10' for several successive years, thus encouraging consumers to monitoring prices on its behalf. Using information obtained when reimbursing consumers, Carrefour systematically asked the relevant suppliers to ‘solve the problem' caused by the lower prices offered by its competitors. »

SSA – 2015 - 54

The Toys Case – 3 –

Companies Fines

(€)

Carrefour France 27 400 000 Maxi Toys France 1 800 000 EPSE-JouéClub 300 000

Hasbro 5 100 000 Lego 1 600 000

Chicco-Puériculture France 600 000

MegaBrands Europe NV 240 000 Goliath 25 000

SSA – 2015 - 55

When can we expect Cartels to be broke up?

•  The game is often given up when one of the companies of the cartel is acquired by an outsider. The new owner notices what’s going on and, to keep from getting mired in it, rings up Brussels.

•  Evolving environment: blowing the whistle must be done quickly, since the members of the cartel often start thinking the same thing at the same time

•  Still, differences between countries...

SSA – 2015 - 56

When can we expect Cartels to be broke up? •  Still, differences between countries.

•  The sole French company is the former Rhône-Poulenc, which denounced two different cartels in the vitamin trade nearly fifteen years ago.

•  In Germany, by contrast, company groups traumatised by the Siemens bribery scandal of ten years ago no longer wait around to blow the whistle.

•  The Italian groups are not conspicuous in their hurry to come clean in this matter. In 2002, Deltafina did the forbidden and betrayed an agreement to fix prices in the transalpine raw tobacco market. Was it afraid of its own boldness? At the next meeting with the members of the cartel, the group warned the others that it had already lost its nerve – a kind of whistle-blowing forbidden by Brussels. As a result, Deltafina lost its immunity and had to pay a fine like the others. Cultural reflexes die hard.

SSA – 2015 - 57

Conclusion: whether the legislation actually works even now, remains a subject of controversy.

•  Let’s come back to facts! •  Cartels may exist with a large number of participants

•  See next slide •  Cartels may stay active for a long period

•  See the TV and computer monitor tubes case (10 years) •  Cartels may be detected and rebirth

•  Many firms are recidivist!

•  Possible Solutions •  Increase the probability for the cartels to be detected

•  Leniency program is a step -- with possible good answers from firms (See amount of fines decreasing in 2011 and 2012)

•  Increase fines! (See next slide) •  Increase non monetary fines – reputation effects •  What else?

SSA – 2015 - 58

SSA – 2015 - 59

Fines vs. Jail

•  “The fines are limited to a maximum of 10 per cent of sales. How little that is, was made clear in 2002 by the exposure of cement cartel. According to calculations of the cartel office, this brought its clients profits of around €2bn. All the same, the companies ended up paying only €400 million in fines.

•  Although cartels do enormous harm, the punishment meted out is at the level of fines for traffic violations. Their collusions are merely misdemeanours. In consequence, none of the perpetrators must appear personally before a court. The public usually does not even learn their names.

•  Things are rather different in the US. There, cartel members face long prison sentences, and in 2004 the maximum imprisonment was even extended to 10 years. Ireland and Britain have followed America's lead. The German federal government, though, does not want to hear of this.”

•  “Reluctance to turn EU antitrust legislation into criminal law”.

SSA – 2015 - 60

Fines vs. Jail

SSA – 2015 - 61

Last: a Clever Way to Detect Cartels (or Whatever Data Manipulation) !

•  Benford Law: refers to the frequency distribution of digits in many (but not all) real-life sources of data.

•  The number 1 occurs as the leading digit about 30% of the time, while larger numbers occur in that position less frequently: 9 as the first digit less than 5% of the time

•  This result has been found to apply to a wide variety of data sets, including electricity bills, street addresses, stock prices, population numbers, death rates, lengths of rivers, … and bids received in call for tenders!

•  Also used in order to detect possible fraud in lists of socio-economic data submitted in support of public planning decisions.

•  Another example: the macroeconomic data the Greek government reported to the European Union before entering the Euro Zone was shown to be probably fraudulent using Benford's Law, albeit years after the country joined.

SSA – 2015 - 62

Benford Law!!!

class3-competition and cartels in europe

Documents