Horizontal Mergers

1
Horizontal Mergers

• Carmo Seabra

2
3.3.3. Mergers (Concentrations)

• We consider mergers between firms acting on the
  same market horizontal mergers if the firms
  operate at the same production stage and vertical
  mergers between firms acting at different stages.
  Mergers between firms in different markets
  (clothing and beer...) are less likely to afect
  competition and will not be adressed.
• Mergers may have very different sources
• small firms trying to increase their rate of
  growth
• One management group convinced it can enhance the
  value of a competitor
• Firms with complementary assets trying to
  increase their efficiency
• Firms trying to increase their market power

3
3.3.3 Horizontal Mergers

• An horizontal merger should be evaluated on the
  basis of its
• unilateral effects, or reinforcment of its single
  firm dominance and
• pro-collusive effects or or joint dominance
  (coordinated effects) effects.
• We have seen that collusion is always harmful to
  consumers and social welfare.
• In the case of mergers, the implications to
  welfare are less clear in fact, usually a merger
  increases the market power or the dominant
  position of the merging firms, being detrimental
  to consumers and social welfare. But it may also
  happen that it increases efficiency

4
3.3.3 Unilateral effects of horizontal mergers

• Unilateral effects or reinforcement of single
  dominance occurs when the merged firm does not
  coordinate its behaviour with its rivals firms
  behave non-cooperatively, both before and after
  the merger (art 12 of Law 18/2003)
• Unilateral effects are of two types
• Mergers tends to increase the market power of the
  merging firms, leading to an increase in prices
  and a decrease in consumers surplus and welfare.
  This has distributional and allocative negative
  consequences
• This negative effect may be partially or totally
  offset by efficiency gains

5
3.3.3.Increase in market power

• There is a decrease in the nº of participants,
  whose impact on the competitiveness of the market
  depends on the nº of competitors, on the
  dimension of the merging firms, on their
  strategies...
• If firms compete on prices, (Bertrand) the merge
  leads to an increase on the price of the merged
  firm and of its rivals. Consumers welfare
  decreases.
• If firms compete on quantities, (Cournot) then
  the output of the merged firm decreases (it
  raises its price) but the output of the rivals
  increase (reduce their price). However, overall,
  the effect on consumers surpluse is also negative

6
3.3.3.Increase in market power

• As to profits, in general the merger should
  increase the profits of the merging firms, and
• The profits of outside firms do increase, too. In
  fact, outsiders can gain even more than the
  merging firms themselves. (That is probably the
  reason why outsiders rarely complain of projected
  mergers)
• Producers surplus unambiguously increases
• However, it can be shown that welfare decreases,
  if there are no efficiency gains associated with
  the merger

7
3.3.3. Increase in unilateral market power

• Concentration
• the larger the of firms after the merger, the
  less detrimental it is
• The lower the market shares of merging firms, the
  less detrimental to welfare the merger is and
  even without efficiency gains, a merger between
  two small firms may increase welfare. But mergers
  that increase the size of the largest firm or
  that create a new largest firm always decrease
  welfare
• The relative productive capacities of all the
  participants if the rivals are not able to
  satisfy additional demand, the market power of
  the merged firm increases. Otherwise it
  decreases.

8
3.3.3. Increase in unilateral market power

• - Remember the Concentration Index of
  Herfindahl-Hirschman (HHI)
• More than the level of concentration, one should
  look at its change US agencies consider that
  mergers leading to values of HHI lower than 1000
  do not harm competition.
• If post merger HHI is between 1000 and 1800
  (moderate concentration), the merger is allowed
  unless the increase in the value of HHI is larger
  than 100.

9
3.3.3. Increase in unilateral market power

• If the value of the HHI exceeds 1800 after the
  merger, this will be allowed iif this values
  increase is not larger than 50.
• Another indicator of the effect on competition
  of a merger between firms A and B is the change
  in the elasticity of the residual demand function
  e A,A e A,B this measures the effect on
  Ademand of a 1 percent increase in As price
  minus the effect on As demand of the same
  increase in the price of B. This difference
  expresses the market power of A, ie, when A and B
  coordinate their strategy and increase prices
  simultaneously, firm A will loose the consumers
  going to other firms minus those that would have
  gone to firm B

10
3.3.3. Increase in unilateral market power

• Another relevant indicator of the market power of
  the merging firms is the diversion ration
  dAeAB/eAA the proportion of lost sales that is
  captured by product B when As prices increase,
  or the proportion of consumers that consider B
  their second best alternative to A
• This ratio measures the intensity of pre-merger
  competition between A and B a small value of d
  means that A and B are not close competitors and
  the merge should not have a significant effect on
  market power
• Under some conditions dAmB/1-mA meaning that
  the probability that B is the second best option
  to A equals the market share of B relative to the
  market share of all the participants with the
  exception of A

11
3.3.3. Other factors affecting the unilateral
impact of a merger

• Entry the easier and less costly entry is, the
  lower the market power enjoyed by merging firms
• Potential entrants that anticipated negative
  profits at the pre-merger price may become
  profitable after the merger this possibility may
  restrain the merging firm from increasing prices
• The countervailing buyers power strong buyers
  limit the sellerss market power
• This is true but does not imply that the lower
  prices are passed to consumers if the buyers
  market is in itself non-competitive ...

12
3.3.3. Other factors affecting the unilateral
impact of a merger

• Switching costs increase market power
• The failing firm defence a merger between two
  firms, one of which was comdemned anyway, does
  not hurt competition
• If production depends on the availability of
  essential inputs, the access of rivals to those
  inputs is relevant.
• Techonological innovation
• The structure of the distribution networks

13
3.3.3. Efficiency Gains

• Efficiency gains may outweight the increase in
  market power of merging firms and result in
  higher welfare (lower prices). These may be due
  to
• Rationalisation savings from reallocating output
  from one facility to another, without changing
  joint output. (Marginal costs may be different,
  prior to the merger, due to different amounts os
  physical capital, different levels of knowledge,
  different output levels)
• Economies of scale and scope
• long-run economies of scale in production, RD
  and marketing and distribution). Notice that
  exploitation of economies of scale may imply a
  decrease in product diversity

14
3.3.3. Efficiency Gains

• short-run economies of scale (due to elimination
  of indivisibilities in inputs, like billing,
  purchases of materials) this implies a decrease
  in fixed costs.
• Technological Progress either process
  innovations or product innovations or quality
  improvements
• Diffusion of know-how (either one way or two way)
• Incentives for R D (too much competition could
  destroy the incentives for research, and a merger
  coul help the internalisation of benefits

15
3.3.3. Efficiency Gains

• Purchasing Economies
• If factor markets are imperfectly competitive
  the merger can increase welfare
• The merger can have quantity discounts
• Slack
• Disciplining power of the capital market (threat
  of entry and a strict merger policy...)
• When a merger has efficiency gains, and an
  overall positive effect on welfare, outside firms
  will always lose profits. Then, a possible test
  for expected efficiency gains associated with an
  announced merger is to look at competitors shares
  value

16
3.3.3.Efficiency Gains

• if these decrease, economic analist predict an
  increase in efficiency of the merged firms (of
  course, these changes can also be due to wrong
  predictions, to speculation, to external shocks,
  to expectations relative to the competition
  authority decision -if CA is expected to prohibit
  the merger, even when it has efficiency gains,
  rivals' shares might even increase, due to the
  regulatory costs incurred by the merging
  firms...).
• If outside firms complain about the
  anti-competitive effects of a merger, CA should
  treat those complaints with great scepticism
  (they can mean that they anticipate significant
  efficiency gains, that is, that the merger is
  welfare increasing).

17
3.3.3. Efficiency Gains

• Notice that efficiency gains decreasing variable
  costs are more likely to affect welfare through
  price reductions and increases in consumers
  welfare, while efficiency gains affecting only
  fixed costs will not lead to price changes, but
  impact profits only.
• Farrell and Shaphir argue that CA should consider
  only efficiency gains obtained by synergies ,
  intimate integration of the firmss unique, hard
  to trade assets while no-synergies efficiency
  gains, like those derived from mere
  reorganization of production among facilities,
  should be disregarded

18
3.3.3. Efficiency Gains

• A problem with the consideration of efficiency
  gains is that they are usually hard to calculate
  and completely dependent on information
  exclusively hels by the firm
• In assessing efficiency gains of a merger it is
  very important to show that these efficiency
  improvements could not have taken place in the
  absence of the merger
• Efficiency gains should be computed net of costs

19
3.3.3.Pro-collusive effects of horizontal mergers

• A merger may reinforce the incentives for
  collusion, or joint dominance, (coordinated
  effects).
• The effect of efficiency gains in joint dominance
  is more difficult to analyse. In general, if
  efficiency gains result in lower costs or higher
  capacity, incentives to deviate increase (and the
  stability of collusion decreases). But if
  symmetry increases, incentives to form stable
  collusive agreements increase, the overall impact
  on welfare may be negative.

20
3.3.3.Remedies

• The decision of competition Authorities
  concerning mergers can be Yes, No, or Yes But,
  imposing remedies for the merger to be allowed.
• Remedies can be of two broad types
• Structural, which change the ownership of
  property rights, i.e., may oblige the merging
  firms to sell part of their holdings (usually
  applied in network industries where the merger
  may create regional monopolies)
• Behavioural impose some restrictions on the
  behaviour of the merging firms, in particular
  concerning the practice of foreclosure

21
3.3.3. Structural Remedies

• Divestitures
• If the merger is expected to have
  anti-competitive consequences on some
  geographical markets or product markets, the CA
  may order a divestment of assets
• These assets can be acquired by an existing
  competitor or by a new firm, but the CA should
  require that the buyer is enabled with the
  conditions to become an active competitor all
  the elements required ...to act as a viable
  competitor..tangible (.RD, production,
  distribution, sales and marketing activities) and
  intangibles (such as intellectual prperty rights
  and goodwill) assets, personnel, supply and sales
  agreements, customer lists.....and so forth
  (Commission Notice on remedies, page 46)

22
3.3.3. Structural Remedies

• Divestitures
• The authority may also order divestiture of
  assets that do not represent an anti-competitive
  thereat, if there are economies of scale or
  network egffects
• Problems with divestitures
• Incentives of the sellers to decrease the value
  of the divested business, and to choose a
  non-effective competitor
• Informational assymetries between seller and
  buyers the seller does not sell all of the
  inputs required for competition to materialise
• Empirical studies have shown that if the buyer
  must continue to have a relationship with the
  seller, either they will collude or the buyer
  neves becomes an efective competitor
• Buyers that intend to collude in the future will
  be willing to pay more for the assets have a
  larger probability of getting them
• If symmetry increases, the outcome may be
  collusion

23
3.3.3. Behavioural Remedies

• Commitments to guarantee a level-playing field on
  the use of assets, technologies, inputs
  (vertical firewalls, obligation of the merged
  firms to sell a technology or an essential input
  to entrant competitors ...)
• A problem with these remedies is that they
  require a continuous monitoring and enforcement.
  Further, they are usually easy to evade, given
  that the CA lacks sufficiently detailed knowledge
  of the industry

24
3.3.3.Empirical evaluation of mergers

• Most empirical studies analysed the evolution of
  profitability accounting profits or shares
  prices. If these increase, either market power
  increased or there were efficiency gains, or a
  combination of the two. This type of evidence can
  not discriminate between the two effects.
• Additional evidence on consumer prices, on the
  prices of the shares of the competing firms (if
  these increase, reinforcement of market power
  dominated) can help the distinction between these
  effects
• There are also some empirical studies that tried
  to directly quantify efficiency gains

25
3.3.3.Empirical evaluation of mergers

• Limits
• The nº of studies is very small
• The evidence on company performance is
  contradictory
• The majority of the studies concern
  manufacturing, in the 60s and 70s and in the USA

26
3.3.3.Empirical evaluation of mergers

• The empirical evidence on the profitability of
  mergers shows very clearly that the potential
  benefits from mergers are usualy exagerated and
  that its costs are often undervalued the market
  overestimates the ability of bidders to manage
  other companies assets
• Mergers as managerial discipline the idea that
  mergers improve efficiency because less efficient
  managers tend to be substituted for more
  efficient ones has not been clearly proved by
  empirical analysis, too. Further , according to
  Matusaka (93) the market prefers bidders who do
  not replace the management of the acquired
  companies

27
3.3.3.Empirical evaluation of mergers

• Empirical analysis of the mergers on the US
  airline industry between 1985 and 1988 shows that
  merged firms tended to price higher than the
  others, suggesting that eventual inctreases in
  efficiency that might have taken place were
  outweighted by the reinforcment of market power,
  leading to overall welfare decreases.
• Conclusions
• Horizontal mergers seem to increase market power
• There is no support for a general presumption
  that mergers create efficiency gains
• In particular cases, however, there were
  segnificant efficiency gains

28
3.3.3.Empirical evaluation of mergers

• Profit flows studies mergers have but modest
  effects on the profitability of merging firms
• If market power increases then the merger
  increases inefficiencies
• Outsiders may gain more than merging firms dont
  have to reduce quantity and profit from the
  increase in prices
• Studies of share prices The most puzzling result
  of this literature is the fact that, while
  profits decrease, share prices increase. The
  pre-emptive merger motive? The merger decreases
  profits compared to the status quo, but increases
  profits compared with the relevant alternative

29
3.3.3.Empirical evaluation of mergers

• The US Competition authorities have been quite
  strict in analysing mergers requiring
  extraordinarily large efficiency gains to allow
  it when there are potential adverse competitive
  effects.
• The EU Merger Guidelines recognizes the role of
  efficiency gains on the appraisal of a merger,
  but the Commission until now has followed a more
  ambiguous approach. Indeed in same cases,
  efficiency gains were even considered a minus, in
  that they would hurt rivals.
• However, in its present formulation, the approval
  of non competitive mergers on the grounds that
  they could create national champions, like France
  wanted, has not passed.

30
3.3.3.Empirical evaluation of mergers

• Criteria for (not) considering efficiency
  increases
• Redistributive gains should not be considered
  saving in costs due to discounts from suppliers
• Cost savings associated with quantity reductions
  
• If the criteria is consumers surplus, cost
  savings in fixed costs, like administrative
  costs, should not be considered
• Effects on other markets in Germany, the UK, and
  Portugal are considered while US courts do not

31
3.3.3.Empirical evaluation of mergers

• There are 3 basic approaches to the appraisal of
  a merger
• Case-by-case analysis high informational costs
• General presumption model, using general
  structural indicators high uncertainty costs,
  because empirical studies show that particular
  circumstances may have large impact on the result
  of a given concentration
• Sequential approach in a first step, structural
  indicators are used to arrive at an initial
  decision. In a second phase an efficiency
  defence with a a more detailed investigation,
  is allowed

32
3.3.3. Summing-up

• In the analysis of a prospective Merger, the CA
  must evaluate
• Its anti-competitive effects
• Eventual increases in efficiency
• In general, mergers increase the welfare of the
  merging firms, but decrease overall welfare.
• In the appraisal of the effects of a merger,
  attention must be taken with the problems of
  information asymmetry between the merging units
  and the authority
• A possible solution is to look only at the
  effects of the operation on the welfare of
  consumers and outside firms, assuming that its
  effect on the welfare of the proponent is always
  positive

33
Vertical Restraints and Vertical Integration

• Carmo Seabra

34
3.3.3.Vertical Restraints

• Often a product goes through different stages of
  production and comercialization before reaching
  the final consumer
• Vertical restraints are agreements and
  contractual provisions betweeen vertically
  related firms. These may be aimed at
• Reducing transactions costs
• Guarantee availability and security of supply...
• Coordination...
• These vertical restraints can take many different
  forms and in some circumstances they may have
  differentiated impact on welfare

35
3.3.3.Vertical Restraints

• Vertical mergers have in common with vertical
  restraints the fact that they both aim at solving
  coordination problems
• However these two problems receive a different
  treatment under Competition Law. In the EU,
  vertical mergers fall under the Mergers
  Regulation, while vertical restraints can be the
  object of art 81 (agreement between firms ) or of
  art 82 if the firm has a dominant position

36
3.3.3.Vertical Restraints

• Example Consider a manufacturer and a retailer.
• Due to an horizontal externality, the effort of
  the retailer associated with the sale of the
  output tends to be sub-optimal in fact, unless
  the retailer has the monopoly of the sale, his
  investments on the promotion of the good has a
  public good characteristic
• This horizontal externality arises because the
  seller shares with other retailers the benefits
  of its efforts to increase sales, while the costs
  remain concentrated. Unless the manufacturer
  devises a mechanism to increase the benefits of
  the retailer, this ones investment will be
  sub-optimal

37
3.3.3.Vertical Restraints

• In order to lead the retailer to increase its
  efforts on the promotion of its sales, the
  manufacturer may pursue a set of possible
  policies such as
• Give the retailer the exclusive of the sale, so
  that it has incentives to advertise the product
  without the fear of creating externalities to
  rivals
• Practice non-linear pricing (a two-part tariffs
  including a fixed fee for access to the product
  plus a variable (low) price per unit sold this
  scheme is common in several brands of clothes

38
3.3.3.Vertical Restraints

• Other possible policies are
• Give the retailer quantity discounts or
  progressive rebates
• Require the sale of a minimum amount or of a
  maximum amount (in this case not to allow for
  price decreases)
• Lead the retailer not to sell competing brands,
  by exclusivity clauses
• Practice resale price maintenance of several
  forms (RPM) (fixed, price ceiling, price floor)
• Create exclusive territorial schemes
• Adopt selective distribution policies (forbid the
  sale in supermarkets , for fear of decreasing the
  image)

39
3.3.3.Vertical Restraints

• These policies need not be available at the same
  time and their respective usefulness depends on
  the particular circumstances for instance, if
  the final price paid by consumers is not easily
  observable by the manufacturer, then RPM is not
  adequate...
• If arbitrage is easy, these vertical restraints
  may be impossible to set. Arbitrage in turn
  depends on the relative costs of search, on
  transaction costs, on transport costs
• Low transport costs eliminate the efectiveness of
  exclusive territory clauses...
• An alternative is a vertical merger

40
3.3.3.Vertical Integration

• Vertical integration involves the firm
  undertaking itself the various stages of the
  production/distribution process the integrated
  firm substitutes internal operations and
  transfers for what would otherwise be market
  transactions
• Williamson stressed that a firm may have several
  motives other than reinforcment of market power
  to vertically integrate
• Minimize the negative effects of uncertainty
• Reduce the dependency of the firm on outside
  suppliers or buyers, particularly relevant if a
  particular production stage requires specific
  assets
• Easen the solution of conflicts, that remain
  within the firm

41
3.3.3.Vertical Restraints

• The welfare effects of vertical restraints depend
  on whether the vertical relations are intra-brand
  or inter-brand
• Intra-brand competition refers to markets where
  retailers sell the same product or brand
• Vertical restraints affecting intra-brand
  competition are usually welfare-increasing they
  allow a better coordination of the vertical chain
  and do not harm competition
• If there is a monopoly at each stage then
  vertical integration avoids double
  marginalization and is welfare increasing, in
  that it allows the internalization of this
  vertical externality

42
3.3.3.Vertical Restraints

• Double Marginalization and vertical integration
• Demand is qa-p
• Firm U (upstream), the manufacturer, has unit
  production cost of c and sells to a retailer,
  firm D (downstream)
• Firm D buys product from firm U at a cost w. Its
  comercialization cost is equal to zero and sells
  to final consumers at p
• First firm U chooses the wholesale price, w
• Second Firm D chooses p, given w

43
3.3.3.Vertical Restraints

• The problem of the downstream firm, under
  vertical separation, is
• Max PD(p-w)q (p-w).(a-p)
• FOC are p(aw)/2
• (1) q(a-w)/2
• PD(a-w)2/4
• Now, firm U knows that the quantity of q it sells
  depends on w. So it chooses w such as
• Max PU(w-c)(a-w)/2
• FOC is w(ac)/2

44
3.3.3.Vertical Restraints

• Substituting back in (1) one gets
• psep(3ac)/2
• PDsep (a-c)2/16
• PUsep (a-c)2/8
• PDsep PDsep 3(a-c)2/16

45
3.3.3.Vertical Restraints

• Under Vertical integration, firm U sells directly
  to consumers. So it chooses p so that
• Max Pvi(p-c)(a-p)
• FOCgtpvi(ac)/2
• qvi (a-c)/2
• Pvi (a-c)2/4
• gtVertical integration is unambiguously better
• Prices are lower, since agtc or the market would
  not exist
• Quantities are larger, because cltwgt
• Consumer surplus increases
• Profits are larger, too firm U can pay firm D at
  least the profit it has under separation
• Overall welfare must increase

46
3.3.3.Vertical Restraints

• If a vertical merger is not possible, several
  alternative and equivalent vertical restraints
  can solve this vertical externality
• A. Resale price maintenance, RPM
• Imposing the final price ppvi(ac)/2, vertical
  profits are maximized. Their distribution between
  the 2 firms depends on the wholesale price, w. If
  firm U is stronger, it will set wpvi(ac)/2 and
  will keep all the profits. The higher is w, the
  higher the share of profits that goes to the
  upstream firm. The same will hold if the upstream
  firm sets a price ceiling p(ac)/2

47
3.3.3.Vertical Restraints

• B. Quantity Fixing, QF
• Firm U may oblige firm D to buy qvi(a-c)/2, or
  qgtqvi. Again, the level of w, which determines
  the relative distribution of profits, depends on
  the bargaining power of the 2 parties
• C. Franchise Fee, FF
• Firm U makes firm D pay Fwq, and sets wc. Then,
  firm Ds problem is
• Max PDff(p-c)(a-p)-F
• The solution is ppvi(ac)/2, qqvi, because F
  does not affect FOC.
• The distribution of profits depends on F, as PUff
  F, and PDff (a-c)2/4-F. If firm U has all the
  bargaining power it will set F(a-c)2/4

48
3.3.3.Vertical Restraints

• Under uncertainty, this equivalence disappears.
  Suppose that the retailer is risk-averse. Then,
• if uncertainty refers to demand, RPM is a perfect
  insurance mechanism for the retailer, and
• if uncertainty affects costs, non-linear pricing
  improves his situation
• Summing-up
• Under double marginalization -a vertical
  externality- vertical mergers, RPM, QF, FF lead
  to higher welfare
• When competition at the retail level increases,
  the price-margin at this level deceases and
  welfare improves Policies that decrease
  competition at the retail level, like ET or
  exclusive retailers aggravates the double
  marginalization problem

49
3.3.3.Vertical Restraints

• Horizontal Externality
• free riding on the provision of services like
  advertising may lead to the under provision of
  these services, which is sub-optimal for firm U
  and for consumers. Again vertical integrationa
  and some vertical restrictions can help solve the
  problem
• Exclusive territories decresases the probability
  that each consumer visits several shops
• Fix resale prices or setting price floors may
  allow retailers to recoup their investments
• Vertical integration can again solve the problem

50
3.3.3.Vertical Restraints

• Conclusions
• Vertical integration and vertical restraints
  equivalent to it, (which lead to the same
  outcome) reduces prices and increases output
  welfare increases
• Different types of restrictions are equivalent
  hence choice between them should depend on the
  particular circumstances (RPM is ineffective when
  prices are non-observable, ET is irrelevant if
  transport costs are too low...)
• Then competition policy should not prohibit some
  of these practices while allowing the others

51
3.3.3.Vertical Restraints

• Effect upon variety
• Consider now that n is variable
• under vertical integration there will be less
  stores an integrated firm when deciding whether
  or not to open a new store, will consider its
  effect upon the profits of the pre-existing
  stores
• Consumer surplus will be lower, but overall
  welfare may still be higher because the fixed
  cost of the marginal firm is saved. (Too much
  competition can lead to excessive differentiation)

52
3.3.3.Vertical Restraints

• So vertical integration decreases prices, but
  decreases variety, too. But most models conclude
  that, under plausible assumptions regarding
  preferences increasing preference for variety-
  welfare increases
• The main point is that vertical and horizontal
  externalities create an opportunity for welfare
  increasing vertical restraints
• No prohibition per se should exist a rule of
  reason is more appropriate
• When the two types of externalities are present,
  the internalization of both externalities
  requires the use of at least two instruments. An
  example is Exclusive territories ET Franchise
  fee FF

53
3.3.3.Vertical Restraints

• ETFF
• ET eliminates the pecuniary horizontal
  externality. (There is no fear of a rival
  undercutting the price) and the non-linear
  contract FF solves the double marginalization
  problem (ET QF does also solve the vertical
  externality)
• If competition is scarce, vertical restraints can
  also reduce welfare in order to attract new
  customers, a vertically integrated monopolist can
  over-invest in quality, for instance. Infra
  marginal consumers (who do not value extra
  effort) may be harmed and overal consumerss
  welfare may decrease

54
3.3.3.Vertical Restraints

• Other efficiency reasons for vertical restraints
  and vertical mergers
• Quality certification by some retailers has a
  value. If other firms can undercut these firms
  prices, no certification will occur Hence RPM or
  selective distribution policies may be welfare
  increasing
• Exclusive contracts because of free-riding
  between products, can increase the level of
  service by retailers
• Exclusive dealing may also increase competition
• Restraints that promote specific investments
  long-term contracts may avoid opportunistic
  behaviour (a firm jumping out of the relationship
  after the other has done specific investments...

55
3.3.3.Vertical Restraints

• Negative Effects of Welfare Restraints
• The Commitment Problem if vertical restraints
  help a manufacturer to commit to high prices,
  they will be welfare decreasing
• Consider a manufacturer who auctions a franchise.
  He would get the monopoly profit, P, if
  canditates believed they would remain the sole
  sellers in the area. But after having sold that
  right for P, the manufacturer has incentives to
  auction a second license, by P/2 anticipating
  this, no one will offer him P on the 1st stage
• Most Favoured Nation clauses, MFN, make a
  commitment credible. Notice that EC
  anti-discrimination policies are equivalent to
  enforcement of MFN.

56
3.3.3.Vertical Restraints

• Conclusions
• Vertical restraints can deteriorate welfare if
  contracts are non-observable, the monopolist has
  incentives to renegotiate them and anticipating
  this, a retailer would not set a contract, which
  deprives the monopolist of its market power. ET
  and RPM can solve this commitment problem and
  decrease welfare.
• Imposing non discrimination or transparency
  clauses solves the commitment problem

57
3.3.3.Vertical Restraints

• Vertical restraints that affect only intra-brand
  competition are mostly welfare-enhancing they
  can solve the vertical externality problem and
  increase profits. In most cases they will
  increase consumers surplus, too
• However, vertival restraints can decrease
  welfare, if they lead to an excessive production
  of services or when they solve the commitment
  problem (to high prices) of the monopolist
• If the market-power of the firms upstream is not
  large, its effect is not serious

58
3.3.3.Inter-brand competition

• Consider now several manufacturers and several
  retailers. Vertical restraints may play an
  anti-competitive role, detrimental to social
  welfare. Inter brand competition can be affected
  by vertical restrainst in different ways
• 1-. Strategic effects of vertical restraints (it
  may be used to relax competition and increase
  profits)
• 2- It may favour collusion RPM increases
  observability and may be used to sustain
  collusive agreements
• 3- It may foreclose efficient entry or deny
  access to essential inputs
• efficiency (double marginalization avoided) and
  competition may be affected in opposite
  directions and the net effect needs to be
  evaluated on a case by case.

59
3.3.3.Leverage and foreclosure

• Exclusive dealing Since the 50s the Chicago
  school contested the possibility that exclusive
  contracts lead to foreclosure
• Posner and Brock arguments
• Consider 1 incumbent monopolist, a buyer and a
  potential entrant the buyer will not sign an
  exclusive dealing agreement with the incumbent
  unless it expects to gain. It may gain either
  because the incumbent is more efficient than
  potential entrants and sets lower prices or
  because it is compensated by the incumbent.
  Posner and Brock show that there will exist a
  compensation that leads the buyer to accept the
  contract only if the incumbent has lower marginal
  cost than the entrant.

60
3.3.3. Exclusive Dealing

• The largest compensation that the incumbent will
  pay amount to area cipMAC, the lost monopoly
  profit
• the minimum compensation required by the buyer
  will be area cipMAC, the loss in consumer surplus
• gt If exclusive contracts are signed, they must
  entail efficiency gains, and should not be
  prohibited

A
pM
M
B
C
ci
ce
61
3.3.3.Leverage and foreclosure

• Post Chicago School arguments
• Exclusive dealing may have anti-competitive
  effects if by excluding a more efficient entrant
  the incumbent makes profit M and is able to
  increase its market power in other markets (a
  possible reason is to enjoy economies of scope,
  unavailable to the entrant) he could convince the
  buyer to accept the deal and use these to
  foreclosure the market
• Other models stress other possible channels
  whereby a less efficient incumbent can
  foreclosure the market (a large number of
  uncoordinated buyers, for example)

62
3.3.3.Policy implications

• Vertical restraints may be welfare-enhancing
• Thus it is not desirable to have a fixed rule vs
  vertical arranjements a case by case analysis
  (rule of reason) should be used
• Properly designed vertical agreements and
  contracts can lead to the same results as
  vertical mergers.
• This implies that these practices should be seen
  as substitutes and given the same tratment under
  competition law enforcment

63
Predation and other Exclusionary abuses

• Carmo Seabra

64
3.3.3.Exclusionary abuses art 82

• Exclusionary abuses behaviour by dominant firms
  that have a foreclosure effect on the market,
  i.e. which are likely to completely or partially
  deter entry or force the exit of
  rival.(monopolisation in the US, art 82 in
  Europe)
• Foreclusure may
• Discourage entry
• Limit expansion
• Encourage the exit of rivals from the market
• That is, it does not require that exit actually
  takes place

65
3.3.3.Exclusionary abuses

• Exclusion of rivals may be pursued through the
  practice of low prices, with the aim of deterring
  entry, by decreasing rivals anticipated
  revenues.
• Predation can also assume the form of non-pricing
  predatory strategies, pursued by a firm with the
  objective of increasing rivalss costs.
  Over-investment, tying and bundling,
  incompatibility decisions, refusal to supply,
  exclusive dealing are examples of such mechanisms
• This type of non-price predatory behaviour may be
  pursued by non-dominant firms, too

66
3.3.3.Exclusionary abuses

• Foreclosure or predation may assume different
  forms
• Horizontal foreclusure
• Predatory pricing
• Single branding rebates
• Tying and bundling
• Vertical foreclosure
• Refusal to supply
• Examples of equivalent behaviours
• Contractual tying vs. a pricing strategy of high
  stand-alone prices in comparison with low bundled
  price for 2 products
• Contractual exclusive branding vs. high rebates
  in single branding

67
3.3.3.Exclusionary abuses

• Refusal to supply vs. high upstream price and low
  downstram (near to final consumers) prices
  (margin squeeze)
• Predation
• The theory of predatory pricing assumes that a
  dominant firm lowers its prices for a
  sufficiently long period to induce competitors to
  leave the market and/or to deter entry.
• To be efective, the predatory strategy must imply
  significant losses to the rivals
• If the entrants are as efficient as the
  incumbent, then this one will suffer losses, too.
• Notice dominant firms may be kept from lowering
  prices, to avoid problems with the CA)

68
3.3.3.Exclusionary abuses

• For predation to be a profitable strategy,
  incumbents must have a reasonable expectation of
  gaining exploitable market power in the future
• To start a predatory behaviour, incumbents must
  also have conditions to survive to its
  competitors, either through greater cash reserves
  (deeper pockets), easiest access to capital
  markets or cross subsidization from other markets
  or other products.
• In some conditions, predation in one product or
  geographic market can have profitable effects on
  other markets, through the enforcement of the
  reputation of being a tough competitor

69
3.3.3.Exclusionary abuses

• This effect can be enforced by the realization of
  investments in (over)capacity in the period
  immediately before entry occurs excess capacity
  lowers prices and these investments could be seen
  as a credible commitment to fight.
• In short, for the practice of a low price to be a
  predatory strategy, and not the result of fierce
  competition it must be shown that
• The firm is dominant
• It has the intention to exclude rivals
• It is profitable to lower prices, because gains
  in the long rum will outweigh present losses

70
3.3.3.Exclusionary abuses

• According to the theory, price predation, to be
  effective, should be difficult to identify. In
  fact, for it to be a rational strategy, the
  victim must be unsure that predation is occurring
  -otherwise it will not leave the market, knowing
  that prices will increase in the future.
• Because of this, predation is hard to prove
• There are some empirical papers trying to
  investigate predatory pricing behaviour McGee
  (1958) studied the Standard Oil Trust behaviour
  at the turn of the century, Burns studied the
  tobbaco trust in the US (1891-1906) Salop and
  White analysed antitrust litigation . Usually
  these analysis show that price predation is very
  hard to prove

71
3.3.3.Exclusionary abuses

• The theory of predatory pricing has been
  criticized by some authors, that argued that
  predation is an irrational strategy.
• In particular, McGee and Easterbrook argued
  that, given the incumbents market share,
  predation would be more costly to him than to his
  rivals, and that his costs would increase with
  the success of the predatory strategy.
• Their main point is that this fact makes
  predation non- credible to competitors. The
  losses associated with the predatory strategy
  must be weighted against future and uncertain
  gains.

72
3.3.3.Exclusionary abuses

• Those future gains are uncertain because there
  are a of factors that may render the strategy
  of deterring entry unsucessful the entrant may
  have celebrated long-term contracts with
  customers that will not allow it to disappear, or
  it may shut down and waite for the increase in
  prices.... And if the entrant in fact gives up
  entry, in the future other firms may come
  in..(but entry and exit have sunk costs..).
• According to McGee, predatory pricing is
  irrational becauses a merger (if allowed) would
  be a better solution.

73
3.3.3.Exclusionary abuses

• And why wouldnt a small firm, even if
  financially constrained, be able to explain the
  predatory strategy to its lenders, thus obtaining
  the necessary funds?
• However note that for future acquisition, a
  strategy of predatory pricing aimed at lowering
  the acquisition price could be a good solution
  predation for mergers

74
3.3.3.Predation

• Theoretical Explanations for predatory behaviour
  Imperfect Information
• A. Reputation models (for future contacts) an
  incumbent may set prices below cost to signal
  that he is a strong competitor
• B. Signalling models (limit-pricing), an
  incumbent may set prices below cost to to
  manipulate the beliefs of rivals and convince
  them that he is more efficient than he really is
  or to damage the market information available to
  the entrants (signal jamming)
• C. Imperfect Financial Markets models financial
  distress increases the risk, and lenders know
  that the larger the distress the larger is the
  probability that owners and managers will engage
  in excessively risky stategies

75
3.3.3.Exclusionary abuses

• Non-price predation, or raising rivals costs,
  concerns the abuse of judicial and/or
  administrative procedures in order to raise
  rivals costs
• If prices stay constant, the predator gains
  market share, as the rival restricts output. If
  prices increase, the predators margin increases.
• Predation can occur through an abuse of
  governnment procedures, including litigation and
  the misuse of licensing or regulatory
  authorities.
• Creation of standards, government regulations,
  exclusionary rights to inputs or sales network
  (vertical restrictions), the denial of access to
  essential facilities, bundling and tie-in sales...

76
3.3.3.Exclusionary abuses

• In all these cases the dominant firm is able to
  spread its costs over a larger output than its
  rivals increasing disproportionately their costs
• Unlike predatory pricing, non-pricing predation
  does not presuppose a dominant position of the
  predator nor that it develops its activity in
  multiple markets
• Advertising and innovation might also be used
  with a predatory objective.
• Several of the strategies that may be of a
  predatory nature can also be the outcome of a
  vigorous competition from an efficient supplier.
  It is very important to distinguish these
  situations.

77
3.3.3.Exclusionary abuses

• Mistaking predation for competitive pricing
  (false positive) is as problematic as mistaking
  competitive pricing for predation (false
  negative). Both errors create problems and
  inhibit competition
• To prove a predatory pricing strategy it is
  necessary
• To prove that costs are below a certain cost
  standard (marginal cost? Other?)
• To prove the intention to exclude rivals
• To prove the capacity of the predator to recoup
  the losses in the future

78
3.3.3.Tests of Predatory Pricing the Areeda
Turner

• In 1975 Areeda and Turner, proposed, in the
  Harvard Law Review, a test to be used by courts
  on the definition of predatory prices, ie, on the
  proof that prices are below a cost standard
• The test reflects the view of its authors that
  predation is rare and that should not be confused
  with vigorous and healthy competition. In fact we
  will see that under this test predation is hard
  to prove.
• The test aims at protecting only firms that are
  as efficient as the incumbent.
• The test focuses on short-term efficiency, rather
  than long-run.

79
3.3.3.Tests of Predatory Pricing the Areeda
Turner

• Its first standard is marginal-cost, but
  recognizing that it is usually difficult to
  measure marginal costs, they use average variable
  costs as a proxy. (the relevant time period, is
  the duration of the low pricing episode)
• Hence, according to the Areeda Turner test,
  prices at or above reasonably anticipated average
  variable costs should be presumed legal, and
  prices below average variable costs cost should
  be illegal. Prices above full costs should be per
  se legal.
• In addition, costs associated with promotional
  spending designed at meeting competition, should
  not be included .

80
3.3.3.Tests of predatory pricing Posner and ATC

• This test raised a number of criticism ...(this
  rule) produces a defendants paradise, a
  monopolists heaven
• Posner argued that the long-run is the relevant
  period of analysis hence prices should be
  compared with long-run marginal costs. Given the
  unavailability of these, the test should use full
  average cost based on the operators records
• The demonstration of a predatory pricing
  strategy would require the test that prices were
  below ATC, plus the demonstration of predatory
  motivation (sometimes on press statements of
  company officials ...), plus the existence of
  some pre-requisites.

81
3.3.3.Output Restriction rule

• Pre-requisites are an initial showing by the
  plaintiff of the market availability to
  predation the predator works in more markets
  that the plaintiffs, markets are concentrated,
  entry is slow, fringe firms are few, the product
  is homogeneous...
• Williamson argued that predatory pricing should
  be seen as a long run business strategy hence
  the marginal cost-based tests proposed by Areeda,
  although correct, are incomplete, because they do
  not take into account the post-predatory period.

82
3.3.3.Output Restriction rule

• For instance if a dominant firm anticipates that
  entry may occur and that it will be subject to
  the average variable cost rule, it can over
  invest in the period prior to entry, expanding
  output and lowering prices without violating the
  rule.
• These side effects of anti-predatory prices
  should be taken into account when designing
  policy. Otherwise equally efficient entrants
  could be forced out

83
3.3.3.Output Restriction rule

• So he proposed to prohibit dominant firms to
  expand output in the 12 to 18 months after entry
  occurs.. This time should be sufficient for
  entrants to gain experience and customers. During
  this period, price could not fall below average
  variable cost, too. It it did, the firm would
  have to decrease output in order to increase
  price.
• Besides proving that this rule would be better in
  terms of welfare, he also considered that it
  would be easier to implement, because it wouldnt
  require estimating average and marginal costs.
• In general prices should cover long-run average
  costs,

84
3.3.3.Baumol and strategic predation

• While Williamson focused on controlling output
  increases in response to entry, Baumol looked at
  price increases after the entrant has been driven
  out the market. He proposed that any price cut
  made in response to entry should continue after
  exit, for some 5 years. This way, he eliminates
  the incentives of the predator to suffer losses
  in the first period anticipating monopoly profits
  in the second period
• This rule dispenses any cost accounts and would
  allow equally efficient firms to survive.

85
3.3.3. Scherer rule-of reason tests

• Scherer argues that long-run allocative
  efficiency can not be guaranteed by a simple
  short-run cost-based test. He proposes a
  wide-ranging inquiry into the predators conduct
  and intentions.
• He stands that long-run welfare may be maximised
  with prices above or below marginal costs,
  depending on the circumstances that depends on
  the relative efficiencies of the firms, the
  minimum efficient scale...
• This proposal, of a wide ranging inquiry into the
  particular conditions of the market intentions...
  has regained attention in the EU

86
3.3.3.Phlips

• He states that the Williamson and Baumol
  approaches could facilitate collusion, and that
  the Areeda Turner test is under-inclusive and
  difficult to apply he defended the idea of the
  inquiry.
• The benchmark, ie, the normal competitive price
  would be the non-collusive profit maximising
  oligopoly price (The Cournot Nash solution)
• The inquiry should aim at determining wether the
  conduct of the alleged predator had turned a
  positive entry valueinto a negative one

87
3.3.3. Two-tier Rule of Joskow and Klevorick

• Joskow and Klevorick in a 1979 article proposed a
  two stage approach to predatory pricing. The idea
  is that the first thing to do is to assess wether
  predation is a workable strategy in the market
  where it allegedly occurred. This first phase
  should eliminate harassment suits and decrease
  the risks of false positive cases, ie, confusing
  competition with predation.
• This initial analysis should comprise the
  analysis of
• Short-run monopoly power (predators market
  share, larger than 60, and the elasticity of
  demand for its products)

88
3.3.3.Two-tier Rule of Joskow and Klevorick

• Conditions of entry (capital requirements,
  consumers loyalty, learning curves, quality of
  information about the risks of entry)
• Dynamic effects of competitors and entry (stable
  markets pose more problems than fast growing
  markets, incumbents very innovative are less
  problematic, prices responsive to changes in
  demand or supply condition are also less likely
  to induce predatory behaviour)
• If this first phase analysis leads to the
  conclusion that predation is not likely to occur,
  no pricing rule should be applied

89
3.3.3. Two-tier Rule of Joskow and Klevorick

• In the second stage, a broad inquiry into pricing
  should be conducted
• Prices below average variable costs should be
  deemed predatory
• Prices between average variable costs and
  average costs should be presumed predatory
• Prices above total costs are presumed legal
  unless a price cut in response to entry was
  reversed within two years without a cost or
  demand based reason
• The probability of recouping present losses in
  the future should also be taken into consideration

90
3.3.3.Conclusions

• The predator should have some market power before
  predation takes place
• There must be some cost-based measures against
  which prices are evaluated short run or lung
  run? Variable or total cost?
• Most people agree that RD, advertising,
  management and all other cost should be included
  excluded are only capital costs atributable to
  Investment in land, plant or equipment, property
  taxes and depreciation on land.
• When there is excess capacity due to a drop in
  demand, below full cost pricing is to be allowed
• Intention to exclude is also important

91
3.3.3. Empirical anlysis of predatory pricing

• A full treatment is very difficult and there is a
  strong suspicion that these claims are used to
  support competitors more than competition
• Several things are required an assessment of the
  costs and financial resources of the incumbent,
  of entry barriers (tangible and intangible), the
  viability of the entrant, the actual conduct of
  the incumbent. In particular if there are no
  barriers to entry, a predatory pricing strategy
  is not credible
• Recent antitrust experience shows that actual
  cases are relatively rare, or at least are
  difficult to prove. (In the EU there have only
  been a small of Commission decisions on Article
  82

92
3.3.3.The EC analysis of predatory pricing

• The Commission issued in December 2005 a
  discussion paper setting out a framework for the
  analysis of exclusionary abuses under Article 82
  and inviting comments by 31 March 2006.
• Depending upon the results of the public
  consultation, guidelines may eventually emerge .
• The main objective is to give an economic content
  to the discussion of exclusionary abuses,
  recognizing that Application of Article 82 in a
  form-based way may lead to the prohibition of
  behavior undertaken by dominant firms that may be
  pro-competitive for example, it may lead to
  lower prices or increase inter-brand competition.

93
3.3.3.The EC analysis of predatory pricing

• The Commission is giving priority to exclusionary
  abuses, as this type of abuse is seen as
  particularly harmful to competition. It intends
  to review its approach to exploitative abuses,
  such as excessive pricing
• sets out a general framework for analysis of
  exclusionary abuses. This analysis is then
  applied to four of the most common types of
  abuse
• 1) predatory pricing
• 2) single branding and rebates
• 3) tying and bundling and
• 4) refusal to supply.

94
3.3.3.The EC analysis of predatory pricing

• The Commissions approach is based on the
  following benchmarks
• Pricing below average avoidable costs (AAC)
  will lead to a presumption of an abuse and the
  Commission will not need to further justify its
  finding of abuse. The dominant firm may present
  arguments to rebut this presumption, based on
  lack of foreclosure effect, lack of predatory
  intent and inability to recoup losses. Note
  however, that, in most cases, the AAC will
  coincide with the AVC.

95
3.3.3.The EC analysis of predatory pricing

• o

PgtATC no PltAAC Yes Or PltLRAIC yes (to be
used instead of AAC, in sectors with high fixed
costs ATCgtPgtLRAIC ?? losses are recovered, entry
barriers...
MC
ATC
LRAIC
AAC
AVC
96
3.3.3.The EC analysis of predatory pricing

• Pricing above AAC, but below average total costs
  (ATC) Such behavior will be an abuse where
  there is evidence that the pricing is part of a
  strategy to eliminate competition. Both direct
  and indirect evidence will be taken into account
  to evaluate the likelihood of a foreclosure
  effect in view of the scale, duration and
  continuity of the low prices.
• The Commissions position is that it is not
  necessary to show evidence of actual recovery of
  losses incurred by the dominant firm. Barriers to
  entry will normally be sufficiently high to
  presume the ability to recoup.

97
3.3.3. The EC analysis of predatory pricing

• Pricing below long-run average incremental costs
  (LAIC) LAIC (which will be higher than AAC or
  AVC) may be used as a benchmark instead of AAC in
  certain sectors or markets that have recently
  been liberalized, where fixed costs tend to be
  high and variable costs tend to be low.
• Pricing above ATC This is not generally
  considered to be an abuse because a dominant firm
  that prices above its ATC should only eliminate
  its less efficient rivals. An abuse will only
  occur in exceptional circumstances where there is
  a clear exclusionary strategy (for example, by
  several collectively dominant firms).

98
3.3.3.The EC analysis of predatory pricing

• Single branding
• a dominant firm which uses single-branding
  agreements in relation to a significant part of
  its customer base will have a foreclosure effect.
  
• The Commission will apply a more effects-based
  approach to single branding, and may consider
  evidence that foreclosure effects will not arise
  for example, if agreements are for a short
  duration or depending on the termination rights
  set out in the agreement. These arguments may
  have particular weight where goods are homogenous
  and other competitors have spare capacity.

99
3.3.3.Empirical anlysis of predatory pricing

• RebatesThe Commission will look at the likely
  foreclosure effects of a rebate program. The key
  question for the Commission is whether an as
  efficient operator would be able to compete in a
  market where a rebate program is in place.
• The Commission distinguishes between conditional
  rebates (those granted as a reward for certain
  purchasing behaviour) and unconditional rebates
  (which are applied regardless of the buyers
  purchasing requirements).
• In relation to conditional purchases, the
  Commission will apply the following analysis

100
3.3.3.Empirical anlysis of predatory pricing

• The Commission will calculate the effective
  price of the products given the level of
  discount given to a buyer. The lower the
  effective price as compared with the average
  price of the dominant supplier, the stronger the
  loyalty-enhancing effect.
• If the effective price is above AAC,