Presumption of Agreement: Shifting the Burden of Proof for Cartel Activities under the 2015 Amendment to the Fair Trade Act of Taiwan

Author: Chia-Heng Seetoo, Jamie C. Yang and Janine Lin

The Fair Trade Act amendment that passed the Legislative Yuan’s review on January 22, 2015 (“2015 Amendment”) brought substantive changes to the rules on cartel, merger control review and other areas of the Taiwan antitrust law enforcement. The most significant change in the amended rules is shifting the burden of proof for cartel activity from the authority in charge, the Taiwan Fair Trade Commission (the “TFTC”), to those accused of involvement in cartel activities.

I. The Presumption of Mutual Understanding (Agreement) Introduced by the 2015 Amendment

The expression regarding the definitions of the cartel activities (or concerted actions) and the “mutual understanding” in Paragraphs 1 and 2 of Article 14 of the newly amended Fair Trade Act are slightly adjusted without any changes in its substance. The most notable change, however, is the new Paragraph 3, which provided that “mutual understanding for cartel [among the competing firms] may be presumed by substantially reliable factors such as market situation, characteristics of products or services in question, costs and profits, and the economic rationality of a business’s conducts.

II. The Background of the 2015 Amendment

Finding evidence on the “mutual understanding” between the parties is concededly the most difficult task of enforcement against cartel in any jurisdiction. Because the TFTC has not been equipped with criminal enforcement powers such as search and seizure as in other jurisdictions, it adds another layer of difficulty in securing direct evidences for mutual understanding. The evidentiary struggle that the TFTC went through in the two recent Supreme Administrative Court decisions, 103 Pan-Zhi No. 294 (the “Milk Case”) and 103 Pan-Zhi No. 195 (the “Brewed Coffee Case”) may have prompted the passing of the 2015 Amendment by the Legislative Yuan.

1. The Brewed Coffee Case – April 18, 2014

The TFTC alleged that four major convenience store chains raised their prices of brewed coffee with milk by NT$5 on or around October 2011, and three of the four chains simultaneously launched a campaign of “buy one get another at half-price.” The TFTC fined the four chains for cartel activities, finding no objective supply/demand change that could reasonably explain the chain stores’ raising their prices on brewed coffee.

The Supreme Administrative Court upheld the lower court’s decision, which recognized that “conscious parallel pricing” or “the act of price following” often exist in oligopoly markets.

There is no “meeting of the minds” in these scenarios, and therefore lacking the “mutual understanding” element in a cartel. The Court found that the chain stores’ raising prices could be explained by “conscious parallel pricing” or “the act of price following,” and that the TFTC failed to prove its case.

2. The Milk Case – June 12, 2014

In Taiwan, the fresh milk product market is dominated by three producers, which have a combined market share of over 80%. The fresh milk producers buys from local dairy farms, and sell to consumers through retail channels after processing and packaging.

The TFTC alleged that three major producers increased the price of fresh milk by NT$6 per liter in October 2011, in a concerted manner, and these three producers could not provide sufficient evidence to support their position that pricing decisions were made independently and reasonably. The price of raw milk increased by NT$2, and each of the three producers offered different reasons that contributed to the additional NT$4 increase. These different justifications include costs of raw materials, packaging and overhead; a 1:3 pricing formula based on prior experience; and advertising and sales costs.

The three producers’ above reasons were not accepted by the TFTC because: (1) if underlying justifications for different firms are different, then it is impossible that all three producers reached the same NT$6 increase in the retail price without collusion; (2) the three producers have the motive to fix the price due to the companies’ large combined market shares, inelasticity of fresh milk, and stable conditions of the fresh milk market; and (3) the price increase happened in winter season, a season traditionally has seen lower sales volume compared to summer seasons, and a reasonable producer would not drastically increase the retail price when they face a low-volume sales season, therefore the Commission determined that the price increase did not further the three producer’s economic interest, indicating the existence of collusion among the three producers. Throughout the investigation and subsequent administrative litigation, the TFTC has forwarded no direct evidence of “mutual understanding.”

The Supreme Administrative Court held that “if it could be deduced through analysis of indirect evidence, that it would be impossible to reasonably explain conforming acts in the market but for cartel activities, then it could be deduced that ‘mutual understanding’ of cartel exists. . . . that is to say, if most enterprises in the market make the same adjustments to their prices at the same time, but there is no change in the objective supply/demand on the market, then it can be reasonably presumed that there is a ‘mutual understanding’ between the enterprises for that price adjustment.”

III. The Impact of the Presumption of Mutual Understanding (Agreement)

The new Article 14 paragraph 3 of the Fair Trade Act is seen to have rebutted the outcome in the Brewed Coffee Case, therefore the amendment is sometimes referred to as the “Convenience Store Rule.” The Amendment has not only embraced the decision in the Milk Case but also created a presumption rule with very little safeguard, which is unseen in most modern jurisdictions.

Read literally, the Convenience Store Rule provides that if the TFTC could simply point to any of the above economic factors supported by indirect evidence, the burden of proving nonexistence of cartel activities would be on the enterprises.

The Convenience Store Rule is at odds with most cartel jurisprudence on two main points. First, although the laws in other jurisdictions may recognize strong inferences from circumstantial evidence (such as the “plus factors” in the U.S. antitrust jurisprudence ), but the law enforcement agency or other persons alleging violations will bear the burden of proof. In fact, in recognizing the difficulty in meeting this burden of proof, the E.U. Regulation 1/2003 confers broad powers of enforcement and investigation onto the Commission to aid it in the task of gathering the evidence it requires to establish a breach. Secondly, the vaguely worded Convenience Store Rule provides little safeguard on the minimum level of evidence the TFTC needs to provide to shift the burden of proof to the accused enterprises. Taiwan could well become a jurisdiction in which all “conscious parallel actions” are seen as cartel until proven otherwise.

IV. Critique and Economic Analysis of the New Evidentiary Requirement for Cartel Activities under the 2015 Amendment

There have been serious policy and theoretical debates in law and economic literature about whether the enforcement agency or the court may presume cartel activities by economic evidence. There are two lines of thoughts. First, some scholars (such as Donald F. Turner) pointed out that even if there were no communication whatsoever between oligopolistic firms, each firm will do its best to collect information on the market, to anticipate the optimal move of other firms given its own move, and to reach its own optimal decision of pricing and quantity produced (the Cournot-Nash equilibrium). Because there was no communication and the law cannot be made to prohibit a firm’s “strategic anticipation” of its competitor’s actions, the best that an antitrust policy can do is to alter the market structure unless the enforcement agency can find real evidence of explicit collusion.

The second thought (promoted by Judge Richard A. Posner) is that, the oligopolistic firms may be very cunning in concealing or destroying evidence of explicit collusion, and economic evidence (such as market concentration, price elasticity, or the existence of trade association, to name a few), may help the enforcement agency to focus its limited resources on those industries which may be prone to the cartel activities, or may even help the courts to determine whether the cartel activities did exist to the detriment to the consumers’ welfare.

The second line of thought, however, did not fully consider all aspects of economic theory and therefore is flawed. First, as pointed out by another judge and renowned antitrust scholar Robert Bork, the oligopolistic pricing model only tells us about how the firms would behave, but not whether the oligopolistic pricing, absent any explicit collusion, will achieve the same result as monopolistic pricing. Additionally, Bork pointed out that based on empirical surveys of economists triggered by old U.S. antitrust law precedents, if a geographical location only had one monopolist in the market, adding a second competitor would much more likely drive the price down.

Second, the entire concept of “tacit collusion” neglects the issue of transaction costs. If the few firms intend to collude, in order for them to understand what needs to be observed, these firms need to negotiate the fixed price, the quantity of output reduced, and how to monitor each firm’s compliance with the price-fixing agreement. These monitoring mechanism must ensure that each participating firm will not give secret discounts or increase the outputs, and how non-compliance would be punished given that such explicit agreement is prohibited by antitrust laws. These are reasons why we still see many explicit collusion cases today, where firms were caught holding secret meetings, exchanging emails, or other clear evidence of communication to form a mutual understanding to fix the market price or to reduce the output.

Third, the use of market concentration indicators neglects the issue of production efficiency as pointed out by Bork. In order for a firm to achieve its maximum production efficiency, it often needs to invest a large sum of fixed costs and to produce a minimum sum of quantity so that its products or services may be provided in a cost-effective way, or the scale of economy. This means, in many industries there will only be a handful of firms. Market concentration indicators may help identify industries prone to collusions, but that does not mean collusions necessarily exist. Therefore, from the economic theoretical perspective, whether the concept of “tacit collusion” really exists or not is highly dubious.

Indeed, law enforcement agencies may have poor understanding about economic theory, or may misapply the economic theory to the so-called indirect evidence. The TFTC’s 2011 investigation of the Milk Case is such an example. The TFTC’s argument is flawed for the following reasons. First, high market concentration rate only means it may be more likely for the firms to collude than in another industry where there are many price-taking competitors, but to infer “motive” from the market structure is rather a long stretch, whether legally or economically. Second, in the administrative litigations, the three milk producers mentioned that the distribution channels need to maintain their own profit margin, while representatives from different distribution channels testified that the three producers set the recommended retail price and the distribution channels merely followed the producers’ recommendations. The TFTC (and the administrative courts) thought this is another evidence inferring collusion, but they misunderstood the above statements. Because the wholesale price of fresh milk is different from, and considerable lower than the retail price, if any one of the three producers merely increases the retail price to the same extent as the price increase of raw milk, the profit margin for the distributor channels will become lower. This would have been a good reason for the distributor channels to refuse such producer to place its fresh milk on shelf, because the distributor channels will have less difficulty to sell the fresh milk at a higher retail price due to the low demand and supply elasticity of fresh milk (which is acknowledged by the TFTC). To transport, store, and to sell fresh milk at retail channels in a timely fashion is a costly effort, and the fresh milk producers (who do not have monopoly power in retail channels) have to accept the equilibrium “price” offered by the retail channels, i.e. maintenance of the profit margin.

Finally, the TFTC neglects to consider, and the milk producers did little to raise the issue of “effective” price received by the end consumers in the investigation and administrative litigation proceedings. The NT$6 price increase was only nominal, and different distribution channels may have different policies about offering discounts to end consumers. In order to ameliorate the impact of the NT$6 increase in the nominal retail price, milk producers may offer discounts or promotions to consumers, and if there is no collusion and the three producers are competing against one another, we may see longer promotion periods and steeper discounts in a large, independent retail channel (such as Costco or Carrefour) and less discounts in channels where a single producer can dominate. Carefully structured empirical studies about the discount policies in different retail channels should be able to identify this. Because the price competition in the form of discounts are not considered, the Supreme Administrative Court finally held against the milk producers largely based on the TFTC’s flawed economic reasoning.

V. Conclusion: A Dangerous New Era of Cartel Enforcement

However lacking in economic foundation the Convenience Store Rule may be, it would soon change the cartel enforcement landscape in Taiwan. The Convenience Store Rule may lead to progressive prosecution by the TFTC, now that the burden of proof is easily shifted to the accused enterprises.

As the Convenience Store Rule provides that “mutual understanding for cartel [among the competing firms] may be presumed by ‘substantially reliable’ factors,” the “substantially reliable” qualification would foreseeably become a highly debated point in future cartel enforcement and litigation. However, because the Court in the Milk Case has set the precedent of allowing the TFTC to prove cartel activity with mere economic indicators as explained above, and there are not many jurisprudence on cartel cases requiring a higher degree of evidence and persuasion, it may well be that the “substantially reliable” factors are in fact not so reliable. One can say that in the future, it will be more critical for a firm in an oligopolistic market to assemble a good legal team – such legal team must have a high level of expertise in administrative litigation, the TFTC enforcement practices, and sound knowledge in economic theory.

If you have any questions about any of the topics discussed above, please contact:

Jamie C. Yang
+886.2.7733.7057
jamie.yang@innovatus.com.tw

This note is intended to be a general summary of the law and does not constitute legal advice. You should consult with counsel to determine applicable legal requirements in a specific fact situation.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *