APPC Homepage

It has been 40 years since the Korea Fair Trade Act(Monopoly Regulation and Fair Trade Act) was enacted in 1980 and enforced in 1981. The development of antitrust policy in Korea may be evaluated in various aspects. But the most important one is ‘rationalization or advancement’, which is to apply the effect-based approach of rule-of-reason in assessing anti-competitiveness and efficiency of concerned practices, not the simplistic form-based approach of interpreting verbalism of related acts and decrees. Significant progress and development have been made over the past 20 years, and concrete evidence can be found in the active use of economic analysis in antitrust enforcement. Even though research achievements in industrial organization and antitrust economics have played an influential role in this area before the period, it was in the early 2000s that economists began to submit economic analysis on behalf of Korea Fair Trade Commission (hereafter referred to as “KFTC”) and tackle antitrust cases.

There are several significant momenta in this development. First, in the early 2000s, a network of economists was formed by the Korea branch of Law and Economics Consulting Group, and economic analysis was actively promoted through fair business practices between law firms and their clients. Second, Mr. Chulkyu Kang, who was in charge of KFTC as the chairperson, the first one with the background of economics professor, incidentally around the time (2003.3~2006.3), paid considerable attention to economic arguments submitted by defendant parties as well as establishing an economic analysis team at KFTC in 2005. Third, multinational corporations that performed economic analyses in antitrust cases for granted, retained economists to represent their interests. A typical example is the Microsoft tying case in 2005, wherein the examiner of KFTC and the economic experts on behalf of Microsoft fiercely confronted on various economic issues such as market definition, anticompetitive effects, and efficiency justifications during two-month-long multiple plenary sessions. Lastly, the Court and the KFTC have begun explicitly referring to economic analyses. In this regard, it is noteworthy that the Seoul High Court carefully evaluated the economic analysis of the two parties(the plaintiff Moohak challenging KFTC’s decision to block the merger, on one hand, and the defendant KFTC and the interested 3rd party Daesun defending KFTC’s decision, on the other hand) in the case of the merger of two local soju producers of Moohak and Daesun. The court defined the relevant geographic market which was an important element of anti-competitiveness of the merger and decided in 2004 after clearly determining the validity of their economic analysis. The High Court decision influenced the KFTC’s decision on the merger of beer company Hite and soju company Jinro in 2006.

Economic analysis has played an important role in antitrust enforcement in the United States and the EU, and is now established as an international standard. Kwoka and White (1989) described the term “ increasingly important role of economics in antitrust since the mid-1970s” as ”Antitrust Revolution” in its editorial book, which introduced various economic analysis by executive agencies and defendants in major antitrust cases of the United States. As evidence of the revolution, they suggested two specific aspects of “the ascendancy of economic analysis in antitrust policy” as follows: First, the DOJ antitrust division and the Federal Trade Commission in the United States used the strong power of in-house economists to dig up problem cases, raise economic problems, and use advanced economic tools to evaluate anti-competitiveness and efficiency of related practices. Second, academic economists majoring in industrial organization and econometrics were more actively involved in actual antitrust cases, either as expert witnesses in Court or as government agencies or private undertakings.

I recently published a book titled Fair Trade Cases and Economic Analyses. The book is comprised of 11 ”major” case studies on four mergers, four abuses of market dominance, and three cartels in Korea during the last 20 years. It is differentiated from the aforementioned Kwoka and White (1989, 2018) on US cases and from Lyons (2009) on EU cases. Both of these volumes were a result of collecting the editors’ choice of major case studies which were written by other authors. My book differs from these volumes in that it is the collection of economic analyses which I myself selected from my own research project studies. Hence, the word choice ”major” reflects my own judgement. Even though the selection may have some subjectivity bias and may suffer from being confined to my own experiences, I chose these cases based on their milestone features and the fact that they introduced new economic tools in South Korean antitrust cases. All the chapters were the results of my previous works with co-authors, and were re-written in order to maintain the consistency of framework in the sequence of background information, exposition of important economic analysis, and reappraisal afterwards. Hence, the volume is both editorial and its own new work at once. In the following exposition, I will briefly introduce the main contents of the book and add some commentary.

Part I is “Merger Cases and Economic Analyses”, which dealt with Moohak-Daesun in 2003, Eland-Carrefour in 2006, Ebay-Gmarket in 2009 and SK Telecomm-CJ Hellovision in 2016.

As far as I know, the first case in which the Court recognized the importance of economic analysis in antitrust cases is Moohak-Daesun. The definition of the relevant geographical market was important evidence of the anti-competitive nature of the merger of two local shochu companies, namely whether the competitive market is located in the local Gyeongnam-Busan area or nationwide. If the relevant market was local, the company had a market share of about 90%. However, the market share is only about 10% nationwide. In this regard, I submitted an economic analysis to the Court on behalf of Daesun, the third interested party against the hostile takeover of Moohakr. In that case, I implemented the critical loss analysis of market definition. Even though the method of critical loss analysis was well-known in US and EU at the time, it was practically the first attempt of applying the method in an antitrust case in Korea. Furthermore, the Court resorted to the economic analysis in making the judgement in support or KFTC decision, and clarified its stance on several controversial issues including the appropriate range of fixed/variable costs, and the relevance of the method in implementing SSNIP test for market definition. The ruling was evaluated as an importance precedent that influenced the KFTC decision such as the Hite-Jinro merger in 2006 and future cases.

Eland-Carrefour led the mergers of major retailers such as Emart-Walmart in 2006, and Homeplus-Homever in 2008. My co-worker and I submitted the research project report to KFTC for Eland. In that case, we conducted an econometric analysis to predict competitive effects of the merger using historical data of new entries in various regions. This analysis was later applied to the Emart-Walmart case. This method, known as a ‘natural experiment’, was originally introduced in US Staples-Office Depot merger case in 1997, and has the advantage of directly predicting the merger effect on prices without resorting to a controversial market definitions. Another contribution of the economic analysis was to propose the method of ‘a union of overlapping circles’ in defining geographic regional markets relevant to mergers among large retail stores, supermarkets and department stores and to clarify the implication of ‘one-stop shopping’ in defining product markets relevant to antitrust cases of retail stores.

Ebay-Gmarket was the combination of companies ranked first and second in the “opean market,” with a market share of almost 90% in 2008. The merger, if defined as an “open market,” would be considered legally anti-competitive pursuant to the law. Surprisingly, however, the KFTC granted temporary behavioral remedies, mentioning the dynamic competition in emerging markets could limit anti-competitive damage in the intermediate to long term. I think that this was the first practically important case where KFTC exercised ‘rational discretion’ in allowing presumably anti-competitive mergers. In this case, my co-workers and I submitted a research project report to KFTC on behalf of Ebay. In this analysis, considering both sides of the platforms, the target product market was defined as an “online shopping market” for both sides, rather than a narrower “open market.” In contrast, the KFTC defined buyers without competitive concerns as “online shopping market” and sellers with anti-competitive concerns as “open markets.” Furthermore, in the context of antitrust economic analysis, a merger simulation was attempted in consideration of the interaction between the two online shopping platforms and the efficiency effects of the merger, but there were no competitive obstacles for both buyers and sellers. This is the first merger simulation to evaluate the competitive effects of mergers in South Korea. Merger simulation is often controversial method of evaluating mergers that avoid an explicit definition of the market. Moreover, we emphasized the dynamic competitiveness of the online shopping market, such as active entry and exit, and relocating the business model of current online players to open markets agreed by the KFTC.

The combination of SK Telecomm and CJ Hellovision comprehensive, encompassing vertical, conglomerate, and horizontal mergers. The main anti-competitive concerns were about the combination of SK Telecomm’s nation-wide IPTV and CJ Hellovision’s CATV in paid broadcasting service market in 23 local areas. The KFTC blocked the deal after an unusual 9-month merger review. The question was whether to consider the geographic market for pay broadcasting nationwide like IPTV or local like CATV. In this case, my co-workers and I submitted a research project report to KFTC and the Ministry of Science, ICT and Future Planning for SK Telecomm. The report discussed the relevance of national geographic market in paid broadcasting services based on the nature of market competition with a transition to digital CATV and the logic of ‘a chain of substitution’ with a nexus of nation-wide IPTV. In addition, we conducted the Upward Pricing Pressure analysis (hereafter referred to as “UPP”) in allegedly problematic local areas. UPP analysis is a recent innovation in antitrust merger evaluation with the same merit of avoiding explicit market definition as merger simulation and the advantage of more practical simplification. The KFTC adopted the basic framework of our UPP analysis, which included modifications on some points and misinterpretations on others, and concluded that the anticompetitive harm caused by the merger was serious enough to guarantee structural remedy of blocking.

Part II of my book is “Abuse of Market-Dominant Position Cases and Economic Analyses”, which dealt with POSCO Supply Refusal in 2001, Intel Rebate in 2008, Naver Vertical Restraint in 2014, and Qualcomm Business Model of Chip Supply and Patent Licensing in 2017.

POSCO Supply Refusal is the landmark case of abuse of market-dominant position in that the Supreme Court set the criteria of ‘unreasonableness’ in cases of abuse of market-dominant position, which should be stricter than in cases of unfair trade practices in 2007. Basically, the crux of the decision is the contention that the anti-competitive effects as well as the intents and purposes of concerned practices should be carefully examined. Ultimately, the Supreme Court overruled the decisions made by KFTC and Seoul High Court which found guilty POSCO’s refusal to supply hot steel coils to Hysco, a Posco’s competitor in the downstream cold steel coil market. The main reason was that the concerned practice did not harm market competition seriously even though it might have resulted in some difficulty in terms of Hysco’s business. In this case, my co-worker and I submitted a research project report to the Court on behalf of POSCO. We estimated international demands for hot steel coils employing the Almost Ideal Demand System econometric method, and conducted critical loss analysis for geographic market definition to find that the relevant market should be enlarged at least to the East Asian Market. As a supplement, we analyzed price time-series such as correlation and Granger-causality test to confirm the co-movement of hot coil prices in Korea, Japan and China. This analysis implies that the Supreme Court might have arrived at the same conclusion of POSCO’s being not guilty of abuse of market-dominant position at the stage of defining relevant market and checking market-dominant position instead of the next stage of evaluating competitive effects.

Intel Rebate was challenged by antitrust agencies around the world in Japan, the United States, the EU and South Korea. Intel provided loyalty rebates for PC OEMs such as Samsung, LG Electronics, and Sambo Computer, on the condition of buying CPU chips exclusively from Intel, allegedly with an attempt to exclude Intel’s competitor AMD in CPU market. In this case, I submitted a research project report to KFTC for Intel. In the case of price-based exclusionary practices, a unit price greater than the average avoidable cost should be considered “safe harbor” in accordance with the antitrust tradition. In the case of conditional rebates, the safe harbor test is tightened in that the ‘effective price’ for contestable portion of involved customer’s purchases, not the price for the whole purchase, should be compared with average cost. Furthermore, the cost of comparison should be the cost of the dominant undertakings, not the cost of the target competitor, to ensure that only competitors “as equally efficient” as the dominant business are protected. This is called as the As Efficient Competitor test (hereafter “AEC”). The KFTC refuted my analysis of the effect price-AEC cost test for conditional rebates, and the Seoul High Court agreed to the KFTC’s decision. However, in 2017, the EU’s Court of Justice ruled in favor of the necessity for AEC testing in Intel cases. Even though this decision does not imply that Intel’s practice should be justified, it is nevertheless encouraging in that it established the principle of an effect-based, rather than a form-based, approach, highlighting the usefulness of careful economic analysis in future cases of abuse of market-dominant position.

Naver Vertical Restraint was the practice of Naver Business Platform’s (hereafter “NBP”) restraining its keyword advertising agencies from transporting the right to manage advertisers to other agencies within the NBP platform. This is not a complete prohibition, but an unbinding limit to transportation due to quota or penalty point policies. It is also notable that transportation within the NBP platform was restricted. This different from Google’s practice of hampering the portability of AdWords advertisers and multihoming across its ad platform. It is a practice in which antitrust agencies in the US and EU raised anti-competitive concerns in 2013. In this regard and on some other issues, my co-workers and I submitted a research project report to KFTC on behalf of Naver. We reviewed the literature of industrial organizations on the economic effects of vertical constraints in terms of possible anti-competitive and efficiency improvements. Then, we argued that the NBP’s policy was to ensure that advertising agencies’ free-riding on others’ efforts of initial set-up and following management of advertisers. It is easier and less expensive for an ad agency to steal other agencies’ incumbent advertisers than competing to solicit new advertisers. Moreover, it was not to exclude NBP’s competitors since all ad agencies were multihoming, and also not to facilitate ad agencies’ cartels since they did not have the bargaining power enough to enforce NBP to adopt such policy. The case ended up with a consent decree between KFTC and Naver, one condition of which was to remove NBP’s restraint on transporting. Hence, we did not have the chance of examining the practice carefully in terms of anti-competitiveness or efficiency.

Under South Korea’s fair trade policy, Qualcomm Business Model of Chip Supply and Patent Licensing is s a remarkable case where the KFTC imposed 1.3 trillion won surcharge, but led antitrust agencies in other countries such as Japan FTC, China NDRC, Taiwan FTC, and US FTC in terms of the coverage of concerned practices and the depth of anti-competitiveness arguments. The KFTC accused Qualcomm’s entire business model of abusing its market dominant position by: (1) refusing to license modem chip competitors its mobile technology patents including SEPs (Standard Essential Patents), (2) tying license with chips for mobile handset OEMs, the so-called “no license-no chip”, (3) imposing mobile OEMs unfair license terms such as comprehensive license, high loyalty rate, and free cross-license grant. Then, the KFTC argued that these three conducts are closely related and strengthened in that they have the effect of strengthening Qualcomm’s dual monopolistic position in licensing and chipset markets. I submitted two research project reports to KFTC and Seoul High court for the third interest party against Qualcomm. In the former report, I analyzed anti-competitiveness of Qualcomm’s three business practices with the benchmark of normal practices without such conducts by employing the framework of raising rivals’ costs. In the latter report, I reviewed the decisions of other countries’ antitrust agencies on Qualcomm’s practices, and criticized Qualcomm’s economic defenses forwarded to KFTC, and which were revealed in the lawsuit. Recently (2019.12.4), Seoul High Court made the decision of mostly upholding KFTC decision such as the amount of surcharge, guiltiness of conducts (1) and (2). The Supreme Court’s final judgement remains to be seen.

Part III is “Cartel Cases and Economic Analyses”, which deals with International Graphite Electrode Cartel in 2002, Bid-Rigging in Military Oil Procurement in 2000, and Flour Cartel in 2006. I will briefly introduce the latter two cases, which include interesting legal issues and economic analyses on cartel damage estimation. The economic analysis was conducted during civil litigation after KFTC decisions on the cartels.

Bid-Rigging in Military Oil Procurement may be regarded as a milestone case in estimating cartel damage in two ways. First, the appraiser appointed by the court, the plaintiff (ministry of defense), and the defendants (oil companies) fiercely debated on specifying a rather sophisticated Difference-in-Difference model (hereafter referred to as “DID”), and estimating it appropriately. Second, the Supreme Court’s judgement provided the principles of guidelines for cartel damage estimation which basically conformed to econometric multiple regression. After the KFTC decision in 2000, it took more than 12 years to resolve the damage caused by the cartel starting from the plaintiff’s complaint in 2001, the First Court decision in 2007, the Appeals Court decision in 2009, the Supreme Court decision in 2011 which reversed and remanded the Appeal Court decision, and finally to two parties’ settlement according to the First Court’s damage estimation in 2013. In this long and winding process, the economic analysis by the legal appraiser has become the subject of controversy in this long and winding process. My co-workers and I submitted a research project report to the First Court on behalf of the defendants. We reviewed the appraisal report and identified a number flaws, including data mistreatment, DID model specification, and estimation methods. In retrospect, I think that this case had both some negative and significant effects in terms of contribution to economic analysis in estimating cartel damages. On the negative side, the court of appeal questioned the usefulness of the economic method because the conflict between economists as legal appraisers and consultants on the plaintiffs and defendants has not been resolved. On the positive side, however, the Supreme Court ultimately found the economic method and established the basic principles of cartel damage estimation, i.e., “excluding other effects resulting from product characteristics affecting price formation, economic conditions, market structure, trade conditions, the change in other economic factors, and the extent.” The most appropriate method to implement this principle is through economic multiple regression based on sound industrial economics.

Flour Cartel is an example of collusion in the market of intermediate goods. This case raises a new issue in estimating cartel damage, i.e., passing-on in addition to price overcharge. In this case, the plaintiff (Samrip, a bread producer) asked the appraiser to estimate the amount of damage caused by the cartel price overcharge. On the other hand, the defendant (flour producers) is expected to estimate the amount of damage to the plaintiff’s lost profits after the plaintiff takes into account the excess price of intermediate goods for the portion of the price increase of the final product. The appraiser and co-workers could estimate overcharging of prices using well- established econometric models of cartel dummy variables, but they had to develop theoretical models for deliveries passing-on and estimate the resulting lost profit of the plaintiff. Passing-on is a controversial legal issue in terms of entitling the status of claimant in the civil lawsuit of antitrust damages. In the U.S., the Supreme Court denied both the passing-on defense of producers of intermediate products and the claimant status of final purchasers in the famous Hanover Shoe/Illinois Brick cases. On the other hand, in the EU, both are allowed in principle, even though there have not been many such cases. Flour Cartel is the first case in Korea where the passing-on defense was tried. The First Court did not accept the passing-on defense explicitly, but accommodated it practically at the stage of “limiting the amount of damages,” and confirmed the estimated lost profit of the plaintiff after taking passing-on into account in 2009. The Court of Appeal upheld the lower court’s decision and the parties did not appeal to the Supreme Court any further.

In conclusion, there are many interesting and important antirust cases in Korea where economic analysis played an important role in promoting Korea Fair Trade policy. The cases I’ve introduced in this paper are limited to those I’ve involved in. I have dealt with them from a personal perspective based on my position in such cases. I hope that other economists will share their antitrust expertise and experience with others. I believe that the Asia-Pacific Competition Form will serve as a network and platform for antitrust law and economic experts in the future.