UNIVERSITY OF GRONINGEN
FACULTY OF ECONOMICS AND BUSINESS
MSc Degree Programme in Economics
EVALUATION OF THE EFFECTS OF MERGER REGULATION
Daniela Mamucevska Fundu student number: 1713833
Supervisor:
Dr. L. Schoonbeek
ACKNOWLEDGEMENTS
This master thesis has been shaped and influenced by the advice and support that I have received from my supervisor Dr. Bert Schoonbeek. Here, I would like to express my gratitude for the help and support that he has given me and that have been very valuable for completing this thesis.
I would also like to thank all the teachers at the MSc of Economics programme for the new insights they gave me in economics.
ABSTRACT
This thesis tries to find out whether merger regulation accomplishes its objectives.The analysis draws on reviews of the relevant academic literature and on the experience of competition authorities in the United States, the European Union, The Netherlands and the United Kingdom. The problem is approached by discussing a number of oligopoly models and models that are used by competition authorities in these countries. The main criteria used to evaluate merger control are: changes in social welfare, changes in consumer welfare and price changes.
Summarizing the results of the analysis, we cannot unambiguously say that merger regulation is always beneficial.
CONTENTS
I. Introduction 6
II. Competition policy and its objectives 8
2.1 Definition of competition policy 8
2.2 Objectives of competition policy 9
2.3 The choice of a welfare standard 10
III. Merger theory 15
3.1 Welfare effects of the merger 15
3.1.1 Cournot competition with homogenous goods 15 3.1.2 Bertrand competition with differentiated goods 16 3.2 The impact of mergers on market power 18
3.3 Pro-collusive effects 20
3.4 Efficiency gains from mergers 21
3.4.1 Efficiency trade-offs 23
IV. Merger regulation 25
4.1 The concept of merger policy 25
4.1.1 Market definition 25
4.1.2 Determination of market shares and concentration levels 27 4.1.3 Anti-competitive effects analysis 29
4.2 Mergers decision-making approaches 30
4.3 The treatment of efficiency gains in merger regulation 35
4.4 Types of errors in merger decisions 37
V. Empirical evidence of the effects of merger regulation in practice 40 5.1 The aim of the ex-post assessment of merger decisions 40 5.2 Measuring the effects of merger regulation 42
5.3.1 Structural models and simulations 49 5.3.1.1 Basic approach to merger simulation 50
I. INTRODUCTION
Protecting the interests of consumers and ensuring that firms have an opportunity to compete in free market conditions are treated as important objectives of competition policy (Motta, 2004). Competition policy in most modern economies typically has three main elements: (i) prohibiting agreements or practices that restrict free trading and competition, (ii) prohibiting anti-competitive practices and abusive behavior by a dominant firm within a market, and (iii) supervising the mergers and acquisitions of firms including joint ventures. Transactions that are considered to threaten competition can be forbidden altogether, or approved under specific conditions.
The objectives of competition policy seem simple and non-controversial. Nevertheless, competition policy is controversial. For instance, Crandall and Winston (2003) found out that during the Great Depression, when competition policy was suspended in the US, prices did not rise.
Merger control is probably the most controversial part of competition policy (Crandall and Winston, 2003). Merger regulation should be designed in such a way that mergers which are not detrimental to welfare will be approved and mergers that reduce welfare will be prohibited. However, it may be hard to detect anti-competitive mergers. Mergers can have various welfare implications. They may increase market power of the merging firms or stimulate pro-collusive behavior on the market. Furthermore, mergers may create efficiency gains (Guler et al., 2003).
One of the problems in the analysis of merger regulation is the choice of using the welfare standard to evaluate the effects of such regulation. Usually, competition authorities prefer consumer welfare as main standard, but economists favor a social welfare standard (that is, the sum of consumer welfare and producer welfare). Also, changes in prices can be considered as a measure for the effects of merger control.
benefits of US merger policy are positive. The available empirical evidence on actual mergers demonstrates that a merger may increase prices and inflict harm on consumer welfare (Stewart and Kim 1993, Werden 2007,Gugler et al. 2003, Ivaldi and Verboven 2001).
A critical concern underlying the motivation for this thesis is the recognition that mergers may not always be beneficial for society. Therefore, the thesis tries to answer the following research question: has merger policy been beneficial for society? In order to answer this question we will discuss and review relevant theoretical and empirical studies on mergers in general and merger control in particular. The focus in this thesis is on horizontal mergers. A horizontal merger is a merger between competitors (Motta, 2004).
II. COMPETITION POLICY AND ITS OBJECTIVES
2.1 Definition of competition policy
Modern competition policy has its origin in the United States antitrust legislation of the late nineteenth century, primarily as a reaction to the formation of the trusts in this economy. Its introduction outside the United States began in Europe after the Second World War, and it has since spread to most developed countries.
Competition is by now almost universally recognized as a most important driving force for the efficient allocation of resources, a sustainable long term economic performance and the improvement of social welfare in general. Protecting the interests of consumers and ensuring that firms have an opportunity to compete in free and fair market conditions are often treated as essential objectives of competition policy. Therefore, competition policy is concerned with a lack of competition, which for example is due to an illegal cartel, an abuse of a dominant position, or a merger that creates anti-competition effects.
Motta (2004) defines competition policy as "the set of policies and laws which ensure that competition in the marketplace is not restricted in such a way as to reduce economic welfare." According to this definition, firms might restrict competition in a way which is not necessarily detrimental to society. In other words, some agreements between firms, which limit competition by other firms, may improve economic welfare, and these agreements should be permitted by the competition authority.
The theory of industrial organization argues that the purpose of competition policy is to promote economic efficiency, where "efficiency" is the maximization of the sum of the discounted present value of consumer surplus and producer surplus. According to this definition, current welfare losses may be acceptable, if the market structure which gives rise to the losses will also generate efficiency in the long run, as long as the prospective benefits are not too delayed and the social discount rate is not too high.
the market, in quality and/or in advertising. Hay (2000) points out that competition is not an end in itself. Promoting competition is not sufficient in itself, if the objective of competition policy is to improve economic efficiency (Jenkinson, 2000, pp.73-101). Economic analysis shows that it is not always the case that "more competition" is better. For example: brand proliferation may generate a welfare loss through excessive expenditures on fixed costs associated with each brand; some forms of advertising may be wasteful, as firms seek to duplicate each other's expenditures in order to avoid market- share losses; firms may compete in physical investment or in R&D, leading to duplication of capacity or innovation expenditures.
Competition policy in most modern economies has three main elements: (1) prohibiting agreements or practices that restrict free trading and competition, (2) prohibiting anti-competitive practices and abusive behavior by a dominant firm, and (3) supervising mergers and acquisitions of firms including joint ventures. Transactions that limit competition can be forbidden altogether, or approved under specific conditions such as an obligation to divest part of the merged firms, or to offer licenses or access to facilities to enable other firms to continue competing.
2.2 Objectives of competition policy
Historical reviews illustrate that competition policy and competition laws are often influenced by social and historical factors, and might respond to quite different objectives. Several jurisdictions around the world attribute additional public interest goals to competition policy, such as the promotion of employment, regional development, inflation control, economic stability, and the protection of small and medium enterprises. For example, most notable is the promotion of market integration as one of the key objectives of EU competition policy. Competition laws might also incorporate objectives such as fairness and equity, forcing firms to behave in a certain way both with respect to customers and to rivals. Another concern is that national governments may use competition policy as an instrument to protect domestic markets or to promote the interests of domestic producers in world markets.
law introduced in the United States at the end of nineteenth century, which was initiated due to the complaints of small firms against the large trusts. The European Commission (EC) seems to have taken the view that small and medium sized enterprises are more dynamic, more likely to innovate and more likely to create employment than large firms. But facts that support these arguments are ambiguous.
The favorable treatment of small firms is not necessarily in contrast with the objective of maximizing economic welfare if it is limited to protecting such firms from the abuse of larger enterprises, or giving them a small advantage to balance the financial and economic power of larger rivals. However, helping small firms to survive when they are not operating at an efficient scale of production is in contrast with economic welfare objectives. Indeed, this would encourage inefficient allocation of resources and could contribute to keeping prices high.
Antitrust legislation in most countries around the world points out that the protection of competition should be intended as protecting primarily consumer welfare. In other words, the protection of competition is typically defined giving absolute priority to the protection of consumer surplus, rather than to total surplus. For instance, EU competition policy aims to protect competition on the market as a means of enhancing consumer welfare and of ensuring an efficient allocation of resources. The objective of US antitrust law is to maximize consumer welfare and promote economic efficiency through the optimal allocation of resources in a competitive market context.
However, economists have showed that an exclusive focus on the protection of consumers may be a rather costly policy for society because it disregards any potentially negative effects it can generate on firms' efficiency and on the welfare of the firms' owners and workers. (Buccirrossi et al. 2006, p.24). Therefore, the growing consensus among economists is that the main focus of competition policy implementation should be on maximizing social welfare.
2.3 The choice of a welfare standard
Social welfare is the standard concept used in economics to measure how well economy performs. It is a measure which aggregates consumer surplus and producer surplus. The surplus of an individual consumer is given by the difference between the price that he is willing to pay for the considered good and the price, which actually he has to pay for it. Consumer welfare is the aggregate measure of the surplus of all consumers. The surplus of an individual firm is the profit it makes by selling the good in question. Producer welfare is the sum of all profits made by all firms. Ceteris paribus, an increase of the price at which goods are sold reduces consumer surplus and increases producer surplus. Social welfare is lowest when the market price equals the monopoly price, and highest when it equals marginal cost of production. Hence, an effective competitive process should generate an allocation of the available resources that maximizes social welfare.
The choice of a welfare standard can affect to a large extent the assessment of a merger. For instance, a merger policy intended to maximize consumer welfare might differ from one intended to maximize total welfare. The former would attach no weight to an increase of producer welfare, whereas the latter would attribute no significance to a redistribution of income. According to the consumer welfare standard, a merger ought to be allowed if the gains in productive efficiency that result from the merger are so substantial as to ensure that the price will decrease. According to the social welfare standard, a merger ought to be allowed if it increases total welfare. This implies that even mergers leading to higher prices should be permitted if the gains realized by producers exceed the losses experienced by consumers.
Besides the fact that both EU and US legislators are concerned with economic efficiency, still the weight on producer surplus in practice is typically smaller than that on consumer surplus. This is particularly evident in the legislation on merger control. In US legislation, for example, a merger that increases market concentration might be forbidden unless it is expected to deliver benefits to consumers in terms of a reduced price or improved quality (US Horizontal Merger Guidelines1). Also in EU Merger Control Regulation2 a proposed merger is not allowed if it significantly impedes effective
1www.usdoj.gov/atr/public/guidelines.
competition. Here competition is considered to be significantly impeded if the merger harms consumers' interests.
In the recent debate on the appropriate welfare standard for antitrust interventions (i.e. consumer surplus or social welfare) a number of economic theories have emerged trying to explain why a consumer welfare standard might perform better in practice (eg. Buccirrossi et al.,2006). For example, Hay (2000) and Chone and Linnemer (2008) emphasize the fact that merging firms enjoy information advantages over the competition authority, and that the adaptation of a consumer welfare standard might offset this asymmetry. Some authors suggest that competition authority officials might be exposed to the lobbying of firms that can offer them a personal reward, and claim that a consumer welfare standard might compensate the predisposition resulting from such lobbying. Another reason invoked in favor of the adaptation of a consumer welfare standard is that market power3 has negative effects both on allocative efficiency and productive efficiency. Less competitive pressure and higher market power may induce firms' slack and waste in their operation.
Proponents of a consumer surplus standard also suggest that it might simplify decisions by the competition authority in mergers cases, since it could limit the analysis to the assessment of the effects on prices. Furthermore, the adoption of a consumer welfare standard by the competition authority may lead firms to attach a higher weight to cost reductions and to propose more efficient mergers, which in turn has a positive effect on social welfare. Therefore, their proponents promote the adoption of a consumer surplus standard, partly because they believe that this is also instrumental to the maximization of total welfare.
The economists (eg. Motta, 2004; Williamson, 1968) who are against a consumer welfare standard point out the fact that it does not take into account the gains made by the merged firms. A literal adaptation of the objective of maximizing consumer surplus would lead to pricing at marginal costs, which would have negative effects on firms' incentives to innovate, invest and introduce new products. Therefore, the competition
3 Market power is defined as the ability of firms profitably to maintain prices above the level corresponding
authority should consider merger proposals in dynamic terms; for example by understanding that helping consumers today might hurt consumers tomorrow. Very often dynamic efficiency has been invoked on as a defense of a softer competition policy. Many economists have argued that, in markets where R&D is crucial, it is necessary to adopt a more lenient competition policy, in particular towards mergers. In such industries a firm's size, the magnitude of its customer base, its expected profits and the financial solidity tend to be important determinants of the firm's ability and willingness to invest in high risk R&D intensive projects with long-run returns.
However, this dynamic efficiency argument, a delicate one in itself, cannot be easily applied when the innovation is obtained through a merger. The reason is that competitors may also have to merge in order to imitate the first innovators and re-establish even competitive conditions. It has been argued that such a process would lead to an extremely concentrated market (Paolo Buccirrossi et al., 2006).
Neven and Röller (2005) have addressed the issue of how institutional settings, such as transparency of the lobbying process and accountability of the competition agency may influence the choice of an appropriate welfare standard. Effectiveness in a lobbying process depends on its transparency and the accountability of the competition authority. With greater transparency lobbying activities have to take indirect routes which are typically less efficient. For instance, lobbying takes place through indirect means like expensive lunches or the promise of lucrative jobs in the private sector.
The outcome of the analysis of Neven and Röller (2005) suggests that neither a social welfare standard nor a consumer surplus standard dominates. Instead, the consumer surplus and social welfare standards give rise to different types of inefficiencies: relatively inefficient mergers (which decrease consumer welfare) are pushed through under a social welfare standard, while relatively efficient mergers (which increase social welfare) are prohibited under the consumer surplus standard.
organizational advantages in lobbying competition authorities which tend to bias decisions in their favor. In this context, interested parties (the merging firms and non-merging firms) provide inducements to the competition authority which are contingent on the outcome of the merger review. Consumers do not lobby the competition authority because they are not well informed about the consequences of proposed mergers and may face prohibitive transaction costs in representing their interests.
The incentive for the competition authority to follow the assigned standard is introduced into the framework through the monitoring system, where the actual outcome is compared with the optimal outcome associated with the given standard. Once a review is initiated, a penalty on the competition authority is imposed whenever its decision is not in line with the assigned standard. The bigger the discrepancy in the outcome, the higher the penalty.
As a general conclusion, Neven and Röller (2005) found that in terms of welfare neither a social welfare standard nor a consumer surplus standard provide the first-best solution. Their model demonstrates that the consumer surplus standard is more attractive relative to a social welfare standard, under the following circumstances: (i) when lobbying is efficient (or the transparency is low), (ii) when the competition authority is less accountable, (iii) when the market share of merged firms is large and (iv) when a marginal increase in merger size is highly profitable.
III. MERGER THEORY
This chapter presents the theoretical implications of horizontal mergers - that is mergers between competitors (Motta, 2004). Three main issues should be considered when discussing the effects of mergers. Firstly, mergers may have unilateral effects. Unilateral effects are defined as the effects of a merger on the ability of the merged firms to exercise more market power. Market power is defined as the ability of a firm to increase its price above some competitive level for a considerable period of time. Secondly, a merger may have pro-collusive effects and create favorable conditions for coordinating conduct in the industry. Thirdly, a merger may also create efficiency gains for the merging firms. So, competition authorities need to compare the increase in market power and potential pro-collusive effects against any benefit the merger may bring in terms of efficiencies gains.
3.1 Welfare effects of the merger
Modeling mergers and analyzing their welfare effect is a difficult task. Usually in economic theory, merger analysis is based on two simple models: models where firms produce a homogeneous good and models with differentiated goods. Both models will be briefly discussed below.
3.1.1 Cournot competition with homogeneous goods
A standard oligopoly model is the Cournot model with a homogeneous good. Consider n identical firms which sell a homogeneous good. Firms have constant returns to scale and there are no capacity constraints. Firms have constant marginal cost c. The inverse demand function is given by:
p = a-Q (3.1)
where p is the industry price, a is a demand parameter and Q is total industry demand. The industry equilibrium quantity, price and profit per firm before merger are given by:
) 1 ( ) ( n c a qb + − = ; ) 1 ( ) ( n c a c pb + − + = ; ⎟ ⎠ ⎞ ⎜ ⎝ ⎛ + − = n c a b 1 2 π (3.2)
Now, consider a merger between m+1 firms. This means that in the industry there will be left n-m firms. Then, the new post-merger equilibrium solutions are:
) 1 ( ) ( m n c a qp − + − = ; ) 1 ( ) ( m n c a c pp − + − + = ; ⎟ ⎠ ⎞ ⎜ ⎝ ⎛ − + − = m n c a p 1 2 π (3.3) where subscript "p" indicates for post-merger values.
Studying the profit functions before and after the merger, it is obvious that πp > πb
for all values of m. This means that outsiders of a merger always gain from the merger. Concerning merging firms' profit, the analysis should answer whether the post-merger profit πp of the new merged firm is larger than (m+1) times the pre-merger profit earned
by the m+1 firms. ⎟ ⎠ ⎞ ⎜ ⎝ ⎛ + − ⎟ ⎠ ⎞ ⎜ ⎝ ⎛ − + − + > n c a m n c a m 1 1 2 2 ) 1 ( (3.4)
Using this model, Salant et al. (1983) showed that a merger is unprofitable for merging parties if less than 80% of the firms in a given industry participate in the merger. This is a consequence of the behavior of the outsiders. In the Cournot equilibrium, the merging firms will choose to contract their post-merger output, but the outsider firms choose to increase their output.
Because pp> pb it is clear that consumer surplus is lower after a merger. Given
lower merging firms' profits and lower consumer surplus, and increased outsiders profits, the competition authority should be suspicious about the proposed merger when a total welfare standard is used. When the competition authority uses a consumer surplus as a welfare standard, then it should prohibit the proposed merger.
3.1.2 Bertrand competition with differentiated goods
An analysis of welfare effects of a merger in the standard homogeneous Bertrand model is straightforward. In this model, firms compete in prices, and industry equilibrium is given by:
p1 = p 2=...pn = c and π1 = π2 =... πn=0 (3.5)
Using a product differentiation model under Bertrand competition, Motta (2004) showed that mergers which do not entail efficiency gains enhance market power and decrease social welfare. We will discuss this model here briefly.
The model consists of "n" single-product firms. Firms' marginal costs are identical and equal to c≥ 0. The demand function of firm i is given by:
⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎣ ⎡ + + − = ∑ = n j j i i p p q n v n 1 ) 1 ( 1 γ γ (3.6)
where qi is the demand of firm i and pi its price, and γ, ν >0 are given parameters.
The pre-merger industry equilibrium price, quantity and profit per firm are given by: n n c v pb= + +γ+−γγ−γ 2 )) 1 ( ( (3.7) ) 2 ( ) )( ( γ γ γ γ − + − + − = n n n n n c v qb (3.8) ) 2 ( ) ( 2 2( ) γ γ π γ γ − + − + − = n n c v n n b (3.9)
Now, consider a merger between two firms. A merger creates a firm with two products. Prices are strategic complements. The post-merger equilibrium values for the prices, quantities and profits of the merging firms (i.e. insiders (I)) and the non-merging firms (i.e.the outsiders (O)), respectively, are given by:
⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ + − + − − + + + − − − + = ) 2 ) 1 ( 3 ) 2 (( 2 ) ) 1 ( ( ) ) 1 ( ( 2 2 ) ) 1 ( ( 2 n n n n n n nv n n n c n n O γ γ γ γ π γ (3.15)
When two firms merge, they take into account the negative effects they impose on each other, and raise their price. The non-merging firms will react by increasing their price as well but not as much as the merging firms. Because the merger increases prices (pI>pb and pO>pb), the consumer surplus after the merger is smaller.
Comparison of the pre-merger (πb) and post-merger (πI ) profit of the merging firms shows that the post-merger profit is higher than the pre-merger profit for each firm. This means that the merger benefits the merging firms. Also, the merger increases the outsiders' profits (πO >πb) because, when the merged firm increases its prices, it reduces
the negative externality which affects all the industry. Concluding, the merger reduces consumer surplus, and increases producer surplus (total profits of insiders and outsiders). It turns out that the net social welfare effect is negative, assuming no efficiency gains.
But, if the merger creates sufficiently large efficiency gains, then it might be that it is welfare improving. So, Motta (2004) proposes competition authorities to consider the claimed efficiencies by the merging parties.
3.2. The impact of mergers on market power
i) The impact of a merger on the merging firms: Consider for example a merger between two firms that produce differentiated goods that are substitutes4 and suppose they compete in prices. The merged firm now produces both products. The merged firm gains a monopoly position on the residual demand5 for these two products, given the others' decisions. Before the merger, they were not only reacting to the others' decisions, but also competing among themselves. After the merger, they remove their own competitive constraint and the merged firm will have an incentive to increase its prices. The extent to which the merged firm will increase its prices depends on the degree of substitution between its products and the products of the competitors. A merger between firms that produce closer substitutes gives an incentive to raise price to a large extent. Also, a merged firm increases prices to a lesser extent if the competitors produce more close substitutes.
ii) The impact of a merger on the other competitors: Basically, this impact depends on the type of market competition (prices versus quantities). Again, the focus of the analysis is a merger between firms that produce substitute goods. When firms compete in prices, then prices are usually considered as strategic complements, which means that an increase in the price of one good will lead competitors to increase the price of their own goods as well. Thus, after the merger, an increase in the merging firms' prices will provoke a positive response from competitors and their prices will rise as well. When firms compete in quantities, then quantities are usually strategic substitutes. This means that a reduction in the output of one firm leads competitors to expand their own outputs. Then, a reduction in the merging firms' outputs provokes an opposite response from the competitors.
Ivaldi et al. (2003) find out that the effect of a merger on prices is much more sensitive to the substitutability of the products in the case of price competition than in the case of quantity competition. Also, they conclude that the strategic nature of prices and quantities (complementarity or substitutability, respectively), may influence the firms' incentives to merge. In the case of price competition, strategic complementarity implies
4 For complement goods, the analysis is reversed. For more information see Ivaldi et al. (2003).
5 Demand for firm's "i" product depends not only on the price sensitivity of customers' demand in question,
that a merger is likely to be profitable, because the merging firms eliminate competition between themselves and induce their competitors to increase their own prices. In contrast, in the case of quantity competition, a merger may be unprofitable for the merging firms in the absence of efficiency gains, since the merger triggers a more aggressive response from the other competitors which may suffice to reduce the profits of the merging firms. Notice that these conclusions were confirmed in the models discussed in section 3.1.
3.3. Pro-collusive effects
A second way in which a merger can influence welfare is when the merger generates more favorable conditions for collusion in the industry. Evaluating this impact is a difficult task in practice.
The extent to which collusion might occur after the merger depends on a number of factors. Motta (2004, p.142-159) made a comprehensive overview of the factors that facilitate collusion. Such factors include: the importance of entry barriers, the existence of information exchange among firms, the presence of multi-market contracts, the regularity and frequency of orders, buyer power, demand elasticity, symmetry among firms, ownership links among competitors, the number of firms in the industry and the existence of clauses such as best-price clauses and retail price maintenance.
Evaluating the impact of a merger on possible collusion often involves a delicate assessment. There is no hard evidence of collusive agreements, since the merger is not yet operative. So, competition authorities should compare the pre-merger and the post-merger situation, and analyze whether collusion may actually occur as a consequence of the merger or not. One main difficulty is that collusion may take many forms. Thus, any attempt to evaluate the potential collusive outcome must be based on an a priori judgment of the particular form it can take in the concrete case. Even if there is some evidence on past collusive practices, one has to account for the fact that firms will change their conduct to accommodate to the new market environment created by the merger.
depend on which factors are more important in each particular case. In the sequel we will not further discuss potential pro-collusive effects of mergers.
3.4. Efficiency gains from mergers
Efficiency gains can take many forms. They may be reached in the short run or in the long run, which may require a different treatment. Some of these gains will be passed on to consumers, either through lower prices or through the improvement of the quality of the products. Other efficiencies, for instance the reduction of fixed costs, will translate only into larger profits. Naturally, there are many different ways in which efficiency gains from a merger can be categorized. Röller et al. (2001) propose five groups of potential efficiency gains from mergers. Their typology is based on the concept of the production function and it consists of the following gains:
- Rationalization of production. This term indicates an optimal allocation of the production levels across the different plants of a firm, i.e. rationalization of production refers to cost savings that may be realized from a merger by shifting production from the plants with a high marginal cost to the plants with a lower marginal cost, without changing the firms' joint production possibilities;
- Economies of scale and scope. A firm is said to have economies of scale when its average cost falls as output increases. Economies of scope generalizes the concept of economies of scale to the case of the multi-product firm. Both types of economies are frequently used as an argument to defend a proposed merger;
- Technological progress. This term includes process and product innovation. An innovation-process reduces the producing costs of existing products, and a product- innovation increases the quality of existing products or leads to new products. Both types of innovation imply an improvement in the firms' joint production possibilities frontier;
-Managerial slack. Large public corporations are characterized by a separation between ownership and control. This introduces the problem of asymmetric information between the shareholders of the firm and the management to whom control is delegated. The management is usually better informed about its projects, and the shareholders may have only limited possibilities to change the management's decisions. To collect all the necessary information is a costly process. A failure by the management to maximize the firm's profit leads to so called internal X-inefficiency. Sometimes, mergers might reduce the managerial slack by improving the administrative techniques used to mitigate the problems caused by the separation of ownership and control. Also, mergers might improve efficiency via the substitution of less able managers with more successful ones. Empirical evidence, however, does not seem to give strong support for this "management discipline" theory,6 among others given alternative defensive mechanisms available to prevent a hostile take-over.
A second typology of efficiency effects of mergers is the distinction between real cost-savings and pecuniary (redistributive) cost-savings. This distinction is important, because only real cost-savings are considered in an efficiency defense. Real cost-savings allow merging firms to produce more, or at a higher quality, from the same amount of inputs. On the other hand, pecuniary cost-savings are reductions in merging firms' costs that do not represent real resource savings. For instance, tax gains from a merger or lower marginal input costs due to enhanced bargaining power against suppliers. Pecuniary efficiencies merely transfer wealth without reducing the resources used to produce the product in question.
A third distinction that is often made in merger analysis is the distinction between savings in fixed costs and savings in variable costs. This is important because savings in variable costs, but not savings in fixed costs, benefit both the merging firms and the consumers. The former type of cost reductions is likely to have a direct impact on the prices, while the latter kind of cost reductions will not modify the price decision of the merged firms. If the competition authority attaches a higher weight on consumer welfare,
6 Slack makes a firm undervalued and lowers the firm's stock price. This may induce another company to
then efficiency gains which are due to savings in variable costs should be looked at more favorably.
A fourth distinction of efficiencies is between firm-level and industry-level efficiency gains. An example of efficiencies at the level of merging firms could be cost savings that result from specialization between plants that are owned by the merged entity. An example of industry level efficiencies is that the merger may affect the R&D incentives for competitors. In US and EU merger control, efficiency considerations refer to firm-level efficiencies. A possible reason for this is that they are easier to verify.
A fifth distinction is between efficiencies in the relevant market and efficiencies in related markets. In principle, competition authorities should also pay attention to the effects of the merger that are realized in related markets. This makes it necessary to weigh the gains for consumers in related markets against the losses for consumers in the market where competition is harmed.
Concluding, competition authorities should carefully assess the potential efficiency gains of a merger. This is a difficult task, since they should try to estimate whether or not these efficiency gains are likely to compensate the higher market power enjoyed by the merging firms.
3.4.1. Efficiency trade-offs
Wiliamson (1968) was probably the first author who studied efficiencies in the context of merger policy. His partial equilibrium model can be used to explain the trade-off between price increases and productive efficiencies in merger cases (see figure 3.1).
Figure 3.1 Trade-off between allocative and productive efficiency
What are the static welfare effects of this merger? If a total welfare standard is used, then the loss in consumer surplus has to be compared to the changes in producer surplus. Pre-merger profits are zero and they increased after the merger. Therefore, total welfare declines by the area D, but increases by the area E. So, the net effect is ambiguous in general.
IV. MERGER REGULATION
4.1. The concept of merger policy
In practice, merger case analyses always follow a similar procedure consisting of three tasks. The first task of the competition authority is to define the relevant market with respect to a proposed merger. The second task is to conduct structural analysis, i.e. the determination and analysis of market shares and concentration levels. The third task in a merger analysis is the investigation of possible anti-competitive effects resulting from a proposed merger, taking into account possible efficiency gains. All three tasks will be briefly presented below.
4.1.1. Market definition
Market definition is necessary to assess market power of merging firms. The relevant market is the set of products and geographical areas to which the products of the merging firms belong. A method that is commonly used by competition authorities7 in order to find the relevant market is the so-called SSNIP test (small but significant non-transitory increase in prices), which is also known as the hypothetical monopolist test (Motta, 2004). This test works in the following way.
Suppose that there exists a hypothetical monopolist that is the only producer of orange juice in Spain. Would this hypothetical monopolist find it profitable to increase the price of orange juice above the current level in a non-transitory way, for example, by 10%?8 If the answer is positive, then that orange juice does not face very high competitive constraints. The monopolistic producer does not lose that much demand to a substitute product, say to apple juice. Therefore, the SSNIP test has found the relevant market, i.e. Spanish orange juice should be considered as a separate market.
7 This method is commonly used in US merger control. With the legislation reforms made in 1997, the
European Commission adopted the SSNIP test.
8 The US merger control refers to a 5% increase in prices. The EC merger regulation refers to a 5-10% price
If the SSNIP test finds that a price increase is unprofitable, this implies that the relevant market is not given by the orange juice only. The test should then consider a wider market, for instance including all apple juice as well. If a price increase is still unprofitable, the market has to be widened further, including for instance all fruit juices. This process continues until a price increase is profitable and a separate market has been found. The SSNIP test can be used to identify both the relevant product market and the geographical market.
The SSNIP test relies on the concept of product substitutability. For example, orange juice and apple juice might be perceived by consumers as substitutes. This is demand substitutability. But there might also be substitutability on the supply side, i.e. if a price rise occurs, then a producer that produces all sorts of fruit juices may decide to switch production to more orange juice in a short period of time. In this case, the competitive constraint would not come from the fact that a considerable part of demand would be addressed to competing products when the prices increases, but rather that the price increase attracts producers that are currently selling some other products.
To define the relevant product market the own-price elasticity of a product is basic information. The price elasticity is defined as the percentage change in the quantity demanded resulting from a one percent increase in the price of that product. If the demand for orange juice decreases by 8% because of a price rise of 5%, it is likely that this price rise is unprofitable.
To understand the competitive constraints exercised by other products on the product under examination for market definition purposes, it is useful to estimate the cross-price elasticities. The cross-price elasticity between for example orange juice and apple juice is defined as the percentage change in the demand for apple juice when the price of orange juice rises by 1%. If the elasticity coefficient is low, it indicates that orange juice and apple juice are not perceived by consumers as substitutes, and this suggests a separate market for orange juice.
this test in itself is not always evidence for a high degree of substitutability. Sometimes, external shocks can affect market prices of many products. For example, in a year in which the weather is very bad and the harvest fails, it is very likely that all fruits will be more expensive, and that the price of all fruit juices will increase.
The SSNIP can be used to determine the relevant geographical market. The test then takes the following form: would a hypothetical Spanish monopoly producer of orange juice find it profitable to increase the price of orange juice, for example, 10%? If the answer is positive, then the geographic market will be defined as Spain. If the answer is negative, because the import of orange juice from Portugal makes such a price rise unprofitable, then the SSNIP test should be repeated on hypothetical monopolists' producers of orange juices from Spain and Portugal, and so on.
Tools that help in identifying the geographical market are usually based on import and transaction cost. Concerning the import, the first question is to determine whether import is small with respect to total domestic consumption in a given region, and the second question is to investigate whether exports are a small part of local production. If the answers are positive, then the relevant geographic market is defined by the given region. Transaction costs compared to the product's price give information on the size of the geographic market. If these costs between two regions for the same product market are high, it is not likely that both regions exercise a competitive constraint on each other.
4.1.2. Determination of market shares and concentration levels
The analysis of market power is closely related to the measurement of market shares held by the merging firms. In this part of the analysis two issues should be addressed: firstly, which measure of market concentration should be used, and secondly, which threshold should be used to indicate that a proposed merger will lead to a large increase of market power.
a proposed merger is potentially anti-competitive. The HHI of a market is calculated by summing the squares of the market shares held by all firms in a given industry:
n
HHI = ∑ mi2, mi - market share of firm i, ( i= 1,2,...n) (4.1) i=1
mi = qi/Q, qi- output of firm i, Q- total industry output (4.2)
The determination of the threshold value of the structural indicators is solved in different ways by different anti-competition legislations, and more or less it is an arbitrary decision. For example, in the EC Merger Control Regulation, mergers with a HHI below 1000 normally do not raise competitive problems. For large values of the HHI (>2000) and a high change in HHI (>150), adverse competition effects are possible. When the change in the HHI is over 250, and the post-merger HHI is between 1000 and 2000, the competition authority is also concerned about anti-competitive effects9. In the US Horizontal Merger Guidelines, mergers with HHI below 1000 usually raise no competition concern. Mergers resulting in a market with moderate HHI levels might be anti-competitive when the difference between the HHIs is over 100. For large values of the HHI (>1800), adverse competition effects are presumed if the change in the HHI is over 100.
Table 4.1 EC and US Horizontal Merger Guidelines for HHI standards
Case Eu* Case US**
HHI post-merger Change in HHI anti-competitive effects HHI post-merger Change in HHI anti-competitive effects < 1000 Irrelevant Unlikely < 1000 Irrelevant Unlikely
< 250 Unlikely <100 Unlikely 1000-2000 >250 Possibly 1000-1800 >100 Possibly <150 Unlikely <50 Unlikely >50 Possibly >2000 >150 Possibly >1800 >100 Presumed
* Source: EC Horizontal Merger Guidelines ** Source: US Horizontal Merger Guidelines
Since the EC has higher thresholds for changes in HHI than the US, the number of cases consequently is significantly larger for the US than for the EC. For instance, in 2006, the number of proposed cases in the US was 176810, out of which 37011 were prohibited, and in the EC there were 35612 proposed cases and none of them was prohibited.
The structural indicators are very important for a competition authority because they can exclude potential anti-competitive effects from a proposed merger at an early stage of the merger investigation. For instance, when the market share of two merging firms is smaller than a certain threshold, say 25 % in total, it is very unlikely that the proposed merger will undermine competition. So, the competition authority should allow the merger in that case.
However, a high market share of firms that intend to merge is not sufficient to conclude that such a merge is detrimental. The final outcome of the analysis depends also on the entry conditions in the industry and the buyer power. For instance, a merged firm with a high market share would not be able to increase its price substantially if entry in the industry is very easy or if there is a strong buyer ready to use countervailing power and to switch to competing suppliers.
4.1.3. Anti-competitive effects analysis
This part of a proposed merger investigation analyzes whether unilateral effects and pro-collusive effects will potentially be a problem. Unilateral effects can be analyzed by looking at the ability of the merging firms to impose higher prices after the merger. Actually, this analysis is an empirical question, requiring market data and empirical techniques. Very often there is a problem with data collection and therefore the analysis remains qualitative, taking into account any factors that influence the market power of the merged firms. If it has been established that a proposed merger will induce high market power, then possible issues like efficiency gains, failing firms13 and/or possible remedies are taken into account. When this is completed, a final decision is taken on whether or not
10 Hart-Scott-Rodino Annual Report, Fiscal Year 2006, FTC and DOJ. 11 Ibidem.
12 European Merger Control-Council Regulation 139/204 - Statistics.
to allow the merger. Mergers, that induce entry, generate efficiencies or concern one or more failing firms, may not be considered as being anti-competitive.
4.2 Mergers decision-making approaches
This section aims to present a framework that may be used in practice to incorporate efficiency gains in the merger analysis. Following current practices, Ivaldi et al.(2003) and Röller et al.(2001) propose three approaches that may be distinguished in merger analysis:
i) the case-by-case approach;
ii) the general presumption approach; and iii) the sequential approach.
The case-by-case approach explicitly analyzes the magnitude and effects of efficiency gains in every single proposed merger case. Implementation of this approach entails very high information costs. Basically, there are two types of information gathering activities. Firstly, it is necessary to quantify the market power effects associated with the merger, and secondly, it is necessary to identify and measure any actual efficiency gains that may be realized.
The second approach relies on general presumptions about the potential effects of mergers. The general presumption approach makes use of general structural indicators, such as market shares or concentration indices, with an implicit recognition of the existence of average efficiency gains from mergers. This approach has the potential problem that there is a lot of variation concerning efficiencies from mergers, in which case the structural indicators are not good predictors of the net benefits from mergers.
evaluate, and therefore, these need only be assessed in those cases where anti-competitive effects might be expected.
De la Mano (2002) proposes a seven-step analysis to evaluate the pro-competitive impact of potential efficiency gains. The analysis complements the "competitive assessment test" of EC Merger Regulation and it consists of the following steps: i) isolate the motives for the merger; ii) identify the nature of competitive interaction in the affected markets; iii) identify the nature of efficiency claims; iv) verify the efficiency claims; v) verify that the claimed efficiency effects are merger - specific; vi) estimate the magnitude, the likelihood and the timing of the efficiencies; and vii) estimate the degree of pass-on of the efficiency gains to consumers. Irrespective of whether an explicit assessment of efficiencies is undertaken any further, steps one and two ought to be addressed in every case. Steps three to six analyze efficiency gains per se. The final step seven deals with the ultimate impact of efficiency gains on consumer welfare (for more information see De la Mano, 2002, p. 40-57).
Werden (1996) and Froeb and Werden (1998) develop a test called the "Compensating Marginal Cost Reduction (CMCR) test" for evaluating the net benefits of a proposed merger. The CMCR test calculates the necessary marginal cost reduction needed to offset the merger's price effects. The weak point of this method is that it is valid only in the short-run, when only marginal costs matter. In any event, the method provides competition authorities with a rough benchmark against which to compare claimed efficiency gains. Additionally, Werden and Froeb developed two distinct tests in order to calculate effects from mergers among sellers of a homogeneous product (Cournot competition) and from mergers of sellers of differentiated products (Bertrand competition). Both tests will be briefly discussed below.
i). Cournot competition with homogeneous products
Froeb and Werden (1998) derived a simple condition for implementing a consumer welfare standard in a Cournot setting. They made a simple calculation in order to determine the marginal cost reduction necessary and sufficient to offset the incentive to restrict output and to prevent a decrease in consumer welfare.
of the relevant market is a main precondition. To compute the demand elasticity, one may follow an econometric approach after gathering historical information on prices and total market demand. As an alternative, one may use available data about the price elasticity of market demand from academic publications.
Under Cournot competition, the standard condition for profit maximization is: − = p c p i) ( e si
for each firm i. (4.3) This equation can be rewritten as
e p s e
ci= ( − i) for each firm i (4.4)
where:
p-the industry price of the homogeneous good;
qi and ci - are the quantity and marginal cost of firm i respectively;
e- is the price elasticity of industry demand, defined to be positive, si- the market share of firm i, si=qi/Q;
Q- is the aggregate industry output.
Froeb and Werden (1998) notice that the first-order condition for profit maximization holds before and after the merger. Thus a merger that does not change price will not change Q or e. In that case, after the merger, the rival firms do not change their output decisions and the output share of the merged firms equals the sum of the pre - merger output shares of the merging firms.
Comparing the post-merger marginal cost (c1
i ) (where the post-merger marginal cost is calculated as an average of the pre-merger marginal costs of the two merging firms weighted by their pre-merger market shares) with the pre-merger marginal cost (co
i ), it follows that the proportionate reduction in marginal cost necessary to restore the pre-merger price is given by:
) ( ) ( 2 2 2 0 1 0 s s s s s s c c c k j k j k j i i i e + − + = − (4.5)
According to (4.5), in the case of Cournot competition, the reduction in marginal cost is quite sensitive to the demand elasticity. A higher demand elasticity (for a given price level) induces a lower necessary marginal cost reduction in order to assure no price increase under Cournot competition (see Froeb and Werden, 1998).
Using the CMCR test of Froeb and Werden (1998), De la Mano (2002), made an analysis of the necessary marginal cost reductions under Cournot competition, given different market shares of the merging firms. De la Mano (2002) concluded that in the case where the merger is leading to a market share below 40% and when there is a relatively high demand elasticity (3 or 4), only a modest marginal cost reduction (e.g. 5%) is required to prevent price increases. Moreover, for a given total post-merger market share, a greater pre-merger asymmetry also might reduce the necessary compensating marginal cost reduction. However, very large cost reductions (e.g. 20%) would be necessary to prevent very large mergers from raising the price. This is illustrated in Table 4.2.
Table 4.2 Compensating marginal cost reduction (CMCR) in % terms
market shares Demand elasticity
Firm i Firm j 0.5 0.75 1 2 3 4 5 10.00 10.00 25.00 15.38 11.11 5.26 3.45 2.56 2.04 10.00 20.00 45.00 22.86 16.00 7.27 4.71 3.48 2.76 10.00 30.00 60.00 30.00 20.00 8.57 5.45 4.00 3.16 10.00 40.00 "-" 39.02 24.24 9.64 6.02 4.37 3.43 10.00 50.00 "-" 52.63 29.41 10.64 6.49 4.67 3.65 20.00 20.00 66.67 36.36 25.00 11.11 7.14 5.26 4.17 20.00 30.00 "-" 48.98 32.43 13.79 8.76 6.42 5.06 20.00 40.00 "-" 64.00 40.00 16.00 10.00 7.27 5.71 20.00 50.00 "-" 85.11 48.78 18.02 11.05 7.97 6.23 30.00 30.00 "-" 66.67 42.86 17.65 11.11 8.11 6.38 30.00 40.00 "-" 87.27 53.33 20.87 12.97 9.41 7.38 30.00 50.00 "-" "-" 65.22 23.81 14.56 10.49 8.20 40.00 40.00 "-" "-" 66.67 25.00 15.38 11.11 8.70 40.00 50.00 "-" "-" 81.63 28.78 17.47 12.54 9.78 50.00 50.00 "-" "-" "-" 33.33 20.00 14.29 11.11 Source: De La Mano 2002.
ii) Bertrand competition with differentiated products
Werden (1996) was using a Bertrand competition model with differentiated products and demonstrated that the compensating marginal cost reduction rises, if the pre-merger profit margin and the diversion ratio increase. The diversion ratio is defined as the share of the sales lost by one product of the merged firm that is recaptured by another when the price of the former increases. The diversion ratio and the pre-merger profit margin are useful devices to measure the effects of differentiated products' mergers on competition and consumer welfare. The intuition behind this comes from the fact that a high pre-merger profit margin reflects high pre-merger market power, and a high diversion ratio reflects intense competition between the merging products.
Let each firm i produce a single product i. For each product define pi, qi, and ci -
as the price, quantity and marginal cost, respectively. The profit margin for product i is given by: i i i i p c p m = − (4.6)
With product differentiation, the effects of a merger on prices may be summarized in terms of own-price elasticities (εii) and cross-price elasticities (εij). The diversion ratio to product j from product i is given by:
dq dq q q d i j i ii j ji ji = − = − ε ε (4.7)
In other words, the diversion ratio shows the proportion of consumers for which product j is a close substitute for the product i. Now, suppose that products i and j are merged into a single firm. If the diversion ratio is very small, then products i and j are no close substitutes and the merger would have no significant impact on the market power of firm i. If the diversion ratio is high, and if firms i and j do not face high competition pressure from other firms' products in the market, then a significant increase in the market power is the likely effect.
maximize the sum of the profits for the two products sold by the merging firms with respect to the prices of both products.
Werden (1996) showed that the marginal cost reduction for each firm i necessary to restore pre-merger prices is positively related to the two pre-merger profit margins and the two diversion ratios, that is:
. ) 1 )( 1 ( 0 0 0 d d m p p d m d d m c ji ij i i j ji j ji ij i − − + = • • (4.8)
As Werden (1996) pointed out, a higher profit margin results from greater product differentiation, while higher diversion ratios can be thought of as a consequence of more intense competition between the merging products. Table 4.3 shows the necessary percentage cost reductions for given values of pre-merger profit margins and diversion ratios.
Table 4.3 Necessary marginal cost reduction in % terms for given values of pre-merger margins and diversion ratios
Pre- merger margins (m) Diversion ratios (d) 0.4 0.5 0.6 0.7 0.05 3.51 5.26 7.89 12.28 0.10 7.41 11.11 16.67 25.93 0.15 11.76 17.65 26.47 41.18 0.20 16.67 25.00 37.50 58.33 0.25 22.22 33.33 50.00 77.78 Source:Werden 1996
4.3. The treatment of efficiency gains in merger regulation
The asymmetry in information may also undermine the credibility of claimed efficiency gains.
Another problem is the trade-off between static and dynamic efficiency. Mergers bring about both static and dynamic efficiency implications that may work in opposite directions, i.e. in the short-run a merger can lead to anti-competitive effects, while at the same time it may improve welfare in the long run. Dynamic efficiencies are inherently more difficult to measure and predict. Mainly because of simplicity, the usual practice by competition authorities is to focus their analysis on the short-term effects of a merger.
The efficiency implications of a merger can be framed either as defense or as offence. When merging firms claim efficiency gains, then this is a case of an "efficiency defense". In this case, the competition authority should try to estimate whether or not these efficiency gains are likely to offset the higher market power enjoyed by the merging firms. In the case of an "efficiency offense", cost savings might give the merging firms a significant competitive advantage. The merged firm might set a very low price and as a result, rival firms will not be able to operate profitably anymore, and they are forced to exit the market. According to this position, the competition authority should prohibit such a proposed merger. However, an "efficiency offence" may have only limited validity. Efficiency gains often result in both higher profits of the merging firms and a higher consumer surplus, because of the lower prices that the firm may charge post-merger. So, social welfare might increase as a result of the merger. Also, an "efficiency offence" justification is not focused on the effects on competition, but on the effect on competitors. This is potentially a wrong approach, because competition policy should protect welfare and not rival firms.
In the old regime, the EC never accepted an efficiency defense. For example, in the Aérospatiale-Alenia/de Havilland case, the EC argued that the claimed efficiencies would not benefit consumers and were not merger-specific. The merger was prohibited with the argument that claimed cost savings would counterfeit an otherwise anti-competitive merger. The negative position towards efficiency claims was leading to wrong decisions and much uncertainty in the practice (see Mota, 2004).
that cost savings should be verifiable, merger-specific and beneficial for consumers14. With the new treatment of efficiencies, the EC reduces the possibility of wrong decisions and increases its transparency. Adapting the "efficiency defense" concept will have a number of practical implications concerning merging firms, competitors, consumers and the competition authority. Merging firms will become more prone to discuss the efficiency enhancing effects of their concentration in a framework were these effects are systematically considered. Moreover, the competition agency should further favor this by providing guidance and a clear opportunity to merging firms to discuss formally the expected pro-competitive effects of the merger. In addition, third parties (comprised of both competitors and consumers) should provide direct and indirect evidence, regarding the efficiency claims of the merging firms. Direct evidence might come from customers' views on the impact of the merger on their welfare, while indirect evidence is likely to arise from changes in competitors' stock market prices, following news on the possible merger taking place.
In the US merger legislation the competition authorities are responsive to arguments based on efficiency gains and cost savings of a merger. The US Merger Guidelines indicate that the "efficiency defense" may apply when efficiencies relied upon are merger-specific, verifiable and measurable and when they outweigh or reverse the merger's potential harm to consumers in the relevant market (see Jones and Sufrin, 2004). The US guidelines also suggest that expected net efficiencies must be bigger the more significant are the competitive risks posed by the mergers. This implies that larger levels of market concentration, higher barriers to entry and a lower market demand elasticity, require correspondingly larger efficiency gains to offset anti-competitive effects of mergers in such markets.
4.4 Types of errors in merger decisions
Competition authorities face the difficult task of computing the net effects of a merger. Sometimes, even for firms it may be difficult to compute their net private
benefits from a merger. The task for competition authorities is even larger, since they may have less knowledge regarding the market in which the firms operate.
Given the difficulties in computing the net effects of a merger, and given the constraints in the process of ex-ante merger evaluation, economists distinguish between two types of errors:
i) Type 1 - accept a merger that has net harmful effects; ii) Type 2 - reject a merger that has net beneficial effects.
The challenge of competition policy is to minimize type 1 errors in those cases where significant net harmful effects are likely, and to minimize type 2 errors in those cases where significant net beneficial effects are likely.
Motta and Vasconcelos (2005) developed a model, which deals with dynamic aspects of mergers. They proposed that in order to reduce the number of wrong decisions, competition authorities should consider future potential changes in the industry caused by a proposed merger. Therefore, their model is an attempt of going beyond a static analysis of the mergers' effects and it explicitly considers the role of antitrust agencies in a dynamic merger game.
Motta and Vasconcelos (2005) analyzed a simple model where a merger increases firms' capacity, which in turn leads to scale economies. When such efficiency gains are very small, then the proposed merger must be blocked. When they are of intermediate importance, rival firms lose competitiveness but continue to operate profitably, resulting in a more efficient outcome for the society. However, a merger might give the two merging firms such important cost-savings, that rivals would be unable to survive in the industry. Under these circumstances, in order to survive, competitors are forced to engage in some predatory practices.
other words, if efficiency gains are to be reaped from a merger, then competitors will respond by merging as well, leading to a positive outcome for society.
V. EMPIRICAL EVIDENCE OF THE EFFECTS OF MERGER REGULATION IN PRACTICE
How well merger policy enforcement accomplishes its objectives in practice is a difficult question. There is little empirical evidence that past interventions have provided much direct benefit to consumers or significantly deterred anti-competitive behavior by firms. Due to confidential reasons, the literature has not been able to use all potential sources of data to analyze the effects of the decisions of competition authorities.
How can we sort out whether the mergers that are blocked are the ones that would have led to anti-competitive outcomes and welfare losses? A blocked merger is never observed and thus its effects cannot be compared directly to what would have happened if the merger had been allowed. This difficulty explains why we could not find any case studies that showed that competition authorities prevented significant welfare losses by blocking a proposed merger.
The purpose of this chapter is to discuss the effects of merger regulation in practice. The analysis will be made by analyzing the effects on social welfare, consumer welfare and prices which deduce the effect of merger control. This analysis is based on reviews of relevant and recent theoretical and empirical studies.
Here we would like to point out that in this section we discuss both empirical studies that explicitly analyze the welfare effects of merger regulation and empirical studies that only examine the effects of mergers per se (i.e. without taking into account merger regulation) The latter studies may provide indirect evidence of the potential gain of merger regulation.
5.1 The aim of the ex-post assessment of merger decisions
effects of the actual decision on social welfare (or consumer welfare), to identify the available alternative decisions, to estimate the level of social welfare (or consumer welfare) in these counterfactuals and to compare it with the level achieved with the actual decision. Once the relevant counterfactuals have been defined, competition authorities can estimate a real and potential social welfare (or consumer welfare) and calculate the effects of the merger decisions with respect to the ultimate economic objective of the merger policy.
Another aim of an ex-post assessment that considers the effects of merger decisions is to improve the decision making process in the field of merger control. Consequently, the assessment should help to avoid errors in the analysis that leads to inappropriate decisions.
In conducting research on this subject two problems emerge. The first problem concerns competition as it is inherently difficult to measure. Whereas the second problem concerns the identification of counterfactuals i.e. what would have happened if some alternative decision had been taken instead.
The first problem concerns the choice of variables for measuring the effects of merger regulation. For instance, the change in social welfare or the change of consumer surplus or the change in prices could all be used as measuring variables. However, the choice of the variable is dependent upon the ultimate objective of merger policy.
The second problem may be a difficult task because there are many alternative decisions that could have been taken by the competition authority. For instance, according to EC merger regulation, the competition authority can impose on the merging firms' remedies that the parties themselves proposed. Given this legal framework, the set of the possible decisions that a competition authority can implement is strongly determined by the behavior of the merging firms. Furthermore, competition authorities can choose between the following options in the determination of relevant counterfactuals:
ii) if the parties propose a set of obligations and conditions, the competition authorities can authorize the merger without imposing any remedy, authorize it imposing the proposed remedies, or block the merger. This is shown in Table 5.1.
Table 5.1: The relevant counterfactuals, given the behavior of the merging firms
Decision Merging firms' behavior Counterfactuals
No remedies offered Prohibition Authorization without
conditions Remedies offered Prohibition
No remedies offered* -
Prohibition Authorization with
conditions Remedies offered Authorization without
conditions No remedies offered Authorization without
conditions Authorization without conditions Prohibition Remedies offered Authorization with conditions * This case is not legally possible
Source: Buccirossi et al. (2006)
Once the relevant counterfactuals have been defined, the assessment of a merger decision requires a comparison of the effects that followed the actual decision with the ones that would have resulted from each of these alternatives.
5.2 Measuring the effects of merger regulation
