Instead, splitting by the dichotomy anti- versus pro-competitive mergers yields particularly interesting and reassuring results: On average, the Commission reports 7. This is exactly what we expect by the design of our regressions if remedies present a rent transfer between merging and rival firms. Dunkel is the German word meaning dark , and dunkel beers typically range in color from amber to dark reddish brown. Dunkel is also widespread in parts of Franconia , for example the Franconian Switzerland , where it has been originally the most common beer type.
The region has a lot of microbreweries , of which many still produce Dunkel. From Wikipedia, the free encyclopedia. Redirected from Dunkel lager. This article is about German dark beers. For other uses, see Dunkel disambiguation. This article needs additional citations for verification. Please help improve this article by adding citations to reliable sources. Unsourced material may be challenged and removed.
Thus, we expect a coefficient of minus one in this case: Reassuringly, we get a significant negative coefficient in case of blocking decisions and it is not significantly different from minus one.
However, we do not always get negative coefficients when other remedies have been applied, casting doubts on their general effectiveness. This finding points to substantial rent transfers from merging firms to their rivals when remedies are ordered without solving the anticompetitive problem.
In fact, some of our results could be explained by the market updating its beliefs about a particular antitrust action, once the uncertainty about the merger investigation is resolved. Our main results about the effectiveness of blocking and the relative ineffectiveness of remedies are not only confirmed but even reinforced.
The paper proceeds as follows. In section 2 we present the institutional background of the EU merger control. In section 3 we shortly describe the related literature on the assessment of antitrust decisions. Section 4 discusses our main methodology and hypotheses and highlights our approach to correct the profitability measures for the market expectation about the antitrust investigation. In section 5, we introduce the data, present the results of the event studies, and discuss some measurement issues.
Section 6 highlights our main results and presents a first set of robustness checks. In section 7 we discuss the methodology based on balance sheet data and present a second set of robustness checks. Finally, section 8 sums up and concludes with some remarks. If necessary, a merger can be referred back to the member states for review.
Commission Regulation EC No. These regulations define the legal steps, which serve to control concentrations between undertakings see figure 2.
Merging parties are obliged to notify their intentions to merge to the Commission when the deal has a community dimension. After receiving notification of the concentration, the Commission has 25 working days to assess whether the concentration is compatible with the common market the so called phase 1.
Notice that the Commission cannot outright block a merger after the phase 1 investigation. In this case, the Commission opens the so-called phase 2, which consists of 90 working days, and during which an in depth investigation is carried out. Generally, the Commission makes use of the entire available time, given the problematic nature of these cases, after which it has to come to a final decision: Looking at figure 2, there are three events, which are important for our empirical analysis.
The other two relevant events are the phase 1 and the phase 2 decision dates, which should help us identify the effect of remedial action, as we will discuss in section 4. The study commissioned by the FTC reviews 35 divestiture orders from through Similarly, at the end of the Directorate General for Competition of the European Commission published an in-house study on merger remedies DG Comp, It reviews the design and implementation of 85 different remedies adopted in 40 decisions of the European Commission between and Also in this case, the analysis is done by means of interviews with the committing parties or sellers, licensors and grantors, the purchasers or buyers, licensees and grantees and the trustees.
While certainly informative, the fact that these divestiture studies only use qualitative information interviews for a small number of cases limits their validity for a more comprehensive sample. Eckbo , Stillman , and Eckbo and Wier are the first papers that use event studies to analyze antitrust decisions. Eckbo and Wier look at horizontal and vertical mergers in mining and manufacturing industries of which 76 were challenged.
Although they find significantly positive abnormal returns for rival firms, they argue that this positive valuation effect may be due to positive information released by the merger: Elzinga , Pfunder, Plaine and Whittemore and Rogowsky use a methodology that is based on classifying ordered remedies as successful, sufficient, deficient, or unsuccessful depending on whether they fulfill certain criteria. There are also a number of studies that use event study methodology to evaluate the effects of mergers but do not analyze antitrust authority decisions, such as Banerjee and Eckard , McGuckin et al.
To separate the market power effect from the information effect, they also estimate abnormal returns to rival firms around the time of an antitrust challenge, however, do not find statistically significant abnormal decreases in the stock prices of rival firms.
Thus, they claim that the mergers in their sample do not raise market power on average. Several follow-up studies tackle the same issue. Schumann conducts an event study analysis of 37 acquisitions that were challenged by the FTC over the period and comes up with the same pattern of abnormal returns as in Eckbo and Eckbo and Wier He reports positive abnormal returns to rivals around the antitrust complaint, which are positive and larger for smaller rivals.
They look at the upstream and downstream product market effects of horizontal mergers and identify the customers, suppliers, and rivals of the merging firms. They report positive abnormal returns to rivals of merging firms around announcements, which range from 0. The antitrust challenge of such mergers, however, does not lead to negative abnormal returns for rivals. Similarly 8 For a criticism of this methodology, see McAfee and Williams See also Comanor , who discusses the problem of remedy in the specific Microsoft case.
They focus on merging firms and do not consider the effect on competitors. In another paper, Aktas et al. They reach this conclusion by showing that the likelihood of an intervention by the EU Commission is higher, whenever the merger is proposed by a bidder from outside the EU and has a negative effect on European rivals. Summing up, the evidence on mergers and merger control decisions is rather mixed.
While most studies find positive effects of the merger for rivals, the interpretation differs. Some authors interpret this as being consistent with the information revelation hypothesis e. Eckbo, ; and Eckbo and Wier, , while other authors interpret it as consistent with market power e. There is no event study explicitly analyzing the effectiveness of ordered remedies.
Studies of remedies on a case-by-case approach point to the superiority of structural over behavioral remedies, but leave doubt about their general effectiveness.
Theoretical arguments underline this. Hypotheses and Methods When firms decide to merge, they potentially generate two externalities on rival firms: In both Cournot and Bertrand with differentiated products type of models, market output declines and prices rise absent efficiency gains see Salant et al. Particularly, rival firms gain since they need not bear the quantity reduction of insiders and nevertheless benefit from the higher prices: The 11 Motta et al. While they in principle favor the use of structural remedies to clear problematic mergers, they also point to information asymmetries and incentive problems, as well as to the increased possibility of pro-collusive effects of divestitures.
Farrell argues that the effectiveness of structural remedies may suffer from inadequate buyers, "over" or miss- fixing and the discounting of merger efficiencies. Cabral also qualifies for the superiority of structural remedies. Rival firms lose from fiercer competition. An effective merger control policy should target only anticompetitive mergers: Moreover, in such mergers, the antitrust authority should be able to reduce the rents stemming from increased market power without destroying rents stemming from increased efficiency.
Therefore, to measure merger control effectiveness we need three steps: We next discuss these three steps. Anticompetitive Mergers For a large class of static oligopoly models, a merger generates unilateral incentives to increase prices for both merging firms and rivals at costs of consumers.
Table 1 lists these four possibilities as well as the optimal incidence of an antitrust action. In this case, the authority should act by imposing the appropriate remedies or by blocking the merger. They show that it holds for Cournot and Bertrand with differentiated goods. There should be no remedies in this case, at least none that also reduce the efficiency gains from the merger. For example, Gugler et al. Within this category, two cases can be distinguished: Here remedies may be considered, although the source of the problem is not one of competition policy but one of inadequate corporate governance.
If industry profits go down, nothing can be said about the anticompetitive effects and necessary remedies. See Röller et al. In these models, profit maximizing firms rationally do unprofitable mergers in order to pre-empt rivals, in those cases where being an insider is more profitable than being an outsider.
Of course, another simple explanation is that unprofitable mergers happen just because managers make evaluation mistakes in a world with imperfect information. Under the null hypothesis of efficient markets, abnormal returns have zero mean and a variance equal to: Nij represents the maximum number of firms in each of the two groups for merger j.
Hence, their cumulative average abnormal returns CAAR around the announcement of the merger should be positive i. The decisions of antitrust authorities can also have two major effects. Either the antitrust action - i. If, for instance, remedies are applied to the right mergers - i. If remedies are not effective, stock markets should not react. The extreme action taken by the Commission, i.
We will pick up on this point in the next section, where we develop an approach to account for this problem. The dummy Cj takes on the value of one, if merger j is cleared without commitments and zero otherwise.
The dummy Oj takes on the value of one and zero otherwise, if merger j is cleared with mainly behavioural commitments, such as terminating existing exclusive agreements, granting access to a necessary infrastructure, or licensing agreements, i. The dummy Dj takes on the value of one and zero otherwise, if merger j is only cleared under the commitment that parts of the combined company are divested structural remedies.
The dummy Bj takes on the value of one and zero otherwise, if merger j is blocked after an in depth analysis of the case. The variables contained in X are exogenous controls. It is important to notice that the direction of causality must run from announcement day abnormal returns to decision period abnormal returns: The b-coefficients measure the degree of market power profit reversion due to the final decision of the EU Commission. That is, if remedies are effective, then profits due to market power at the announcement day should be at least partially dissipated by the final decision and b should be negative.
The higher these profits or abnormal returns are around the announcement day, the larger should be the decline in profits abnormal returns due to the decision.
It should be noted that bB , i. We know that blocking is an effective remedy for an anticompetitive concern. Essentially, this complete profit reversion for both merging and rival firms is due to the fact that blocking restores the pre-merger situation. Otherwise, it could happen since a behavioral remedy or a divestiture are considered as rent transfers from merging firms to rivals without remedying the anticompetitive concern.
Therefore, we expect negative positive b's for merging rival firms. Accounting for Market Expectations Until now we have not considered that an efficient market should also account for the future antitrust decision when reacting to a merger announcement. Yet this seems to be a too strong assumption. Assuming that any effective action remedies or blockings 20 However, one can expect even a coefficient smaller than minus one for merging firms, if the opportunity costs of going through the antitrust investigation are judged to be very high.
The first important date is the phase 1 decision. Of course, if all information is public, then there should be no surprise for the market and, hence, we should not observe any abnormal return around the decision day.
Yet, it seems reasonable to assume that some private information generated during the bargaining process between the Commission and the merging parties is unknown to the market. Moreover, a merger can be prohibited only after a phase 2 investigation. Therefore, the abnormal return around the phase 1 decision for those mergers that went into a phase 2 investigation might also reflect the high costs these firms are expected to pay.
The Data and the Estimated Abnormal Returns Our sample consists of concentrations that have been analyzed by the European Commission in the period Our starting database was developed in Duso, Neven, and Röller Our sample includes almost all phase 2 mergers scrutinized by the EC till the end of , and a randomly drawn sample of phase 1 cases, which run up to June Because of difficulties in identifying competitors or their stock, we end up with 78 phase 2 cases and 89 phase 1 cases for which we have complete information.
We identify different firms involved in several mergers either as merging parties or as rivals. This is a customizable business news and research product that integrates contents from newspapers, newswires, journals, research reports, and web sites. We look at the first rumors about the merger, i.
On the other hand, our measure of abnormal returns might be downward biased since there might still be uncertainty on whether the merger will take place or not. Finally, once firms have been identified, we collect information about their characteristics form two sources. This database provides financial information and computation services to the securities industry worldwide.
The market value of the combined firm rivals is on average 45 7. On average, the Commission reports 7. The majority of the concentrations in our sample Remedies have been imposed in Considering these prohibitions as an extreme type of remedy, we have For the purpose of this study we will consider these two categories - which happen to be those used in the merger remedies guidelines - and blockings.
The selling of shares was imposed in Table 3 reports statistics on cumulative average abnormal returns for merging firms and competitors around various events and using different event windows. We consider a short window from 5 days before to 5 days after the merger, and a long window that goes back 50 days before the event. The size of the effects ranges from 1. This result seems to be in line with the literature. The breakdown across later decisions reveals significant effects for those mergers that are either cleared with no remedies in phase 1 efficiency-enhancing mergers?
Looking at phase 1 decisions, we observe negative CAARs for the merging firms as well as for the rivals. The negative effect stems mainly from those cases where a phase 2 investigation was started: The negative CAARs for the merging firms in that case are on average Similarly, rivals lose from the opening of a phase 2 investigation in the long window up to Interestingly, decisions that either clear mergers in phase 1 with or without remedies in the short window and those decisions that lead to later blockings long window decrease rival firms' market value by a rather large and significant 3.
If presumably efficiency enhancing mergers are cleared without - or with ineffective - remedies, or if presumably market power enhancing mergers are blocked, rival firms lose, which is exactly the pattern we predict and observe. For phase 2 decisions, almost all measures of abnormal returns are statistically insignificant. Nevertheless, looking at the long run CAARs, we observe strong negative, though not significant, effects for the rivals in those mergers that were cleared with remedies The market expected some negative effects at the phase 1 decision day for merging firms, but was surprised that the Commission did not impose remedies.
If there is no information leakage, the right measure would be the abnormal 26 See for instance Andrade et al. Depending on the event window, we estimate average abnormal returns for acquirers in the range between These results are quite similar to those reported by Aktas et al.
Note, however, that in their sample the phase 2 cases are more underrepresented than in ours. However, if we think that some relevant information was present in the market already before this event, then a larger window should more carefully capture the real effects.
Looking at figure 3, which shows time plots of daily cumulative abnormal returns averaged over merging firms and rivals, respectively, for the period from 50 days prior until 5 days after major events, we can get a feeling of how much of a surprise the event under consideration was. For both types of firms, we see an upward drift of abnormal returns beginning some 50 days before the announcement of the merger for merging firms and some 30 days for rivals. Phase 1 decisions — specifically, remedies in phase 1 - seem to come as a relative surprise.
This makes sense, since phase 2 is much longer and attracts much more public attention than phase 1. In fact, the probability that the merger will be blocked or cleared with remedies sharply increases when a merger goes into a phase 2 investigation. Results In this section we present the main results and several robustness checks. We start by using the CAARs not corrected for the probability of antitrust action as a benchmark case.
We then look at several sub-samples to qualify our findings. We finally show that the main results hold 29 We however run all tests also using the longer window. Our results remain mostly unchanged. Yet, differences emerge, which helps to explain some of the previous findings.
Main Regressions Table 4 presents our main regression results.
Copyright © 2017 · All Rights Reserved · Maine Council of Churches