My Market Is Not My Market? -An Observation of the Empirical Test Trend for Market Definition (Taiwan)

Oli Wong

The Fair Trade Law prohibits an enterprise from engaging in any competition restraining act that disrupts or distorts market competition mechanisms.  The regulations concerning restraints on competition cover monopolies, mergers or acquisitions, concerted actions, vertical restrictions and other competition-restricting acts, focusing on the negative impact of an enterprise’s behavior on the relevant markets such as impairment, impediment or exclusion.

A relevant market refers to the area or scope in which an enterprise competes.  Only when the relevant markets are determined is it possible to clarify the competitors, the targets of competition, competitive products, areas of competition, and sources of competitive pressure.  The above factors should be ascertained before it is possible to further determine an enterprise’s market power and the anti-competitiveness of its behavior.

“The starting point in any type of comparing analysis is the definition of the relevant market.” ~ OECD

Fundamental aspects of defining the relevant market

An enterprise faces competitive pressure primarily from three sources: demand substitution on the consumption dimension, supply substitution on the competitor dimension and the potential substitution.  Demand substitution, which is focused on observing the potential reactions of the consumers when it comes to demand changes, is the major item considered in defining a market.  Supply substitution and potential substitution, however, are considered only when market access analysis is performed or characteristics of goods or services are reviewed.

Pursuant to the Principles of the Fair Trade Commission for Handling the Definition of the Relevant Market, when the relevant market is defined, the product market and geographic market are two defining dimensions.  However, if the time factor has an important impact on market supplies and demands, the time factor should also be considered in defining the market scope.

Methods for defining the relevant market

The methods for defining the relevant markets generally consist of qualitative analysis, which is descriptive by nature, and quantitative analysis, which is math-based.   When the relevant market is defined, either qualitative analysis or quantitative analysis as applicable may be selected to conduct the analysis, or both qualitative and quantitative analyses may be used concurrently, based on the circumstances of individual cases such as the industry or product characteristics.  There is no fixed requirement.

Qualitative analysis collects relevant data by using participatory observations and in-depth interviews. A typical qualitative analysis includes concepts such as reasonable interchangeability of use and a small but significant non-transitory increase in price (hereinafter, the “SSNIP”) in the hypothetical monopoly test.  In practice, the qualitative analysis mostly uses written data such as the size of operation, market position, pricing and marketing strategies of an enterprise, coupled with the Standard Industrial Classification of the Republic of China formulated by the Directorate-General of Budget, Accounting and Statistics as another qualitative factor, to define the relevant market.

Quantitative analysis needs sufficient product and market data such as product prices and product sales volume for demand/supply estimation, and the results of the quantitative analysis should be consistent with economic theories and the factual behavior in individual cases.    Quantitative analysis includes the calculation of the cross elasticity between the products per se and their substitutes, while the hypothetical monopoly test specifically uses methods such as critical loss analysis, price correlation analysis, the Granger causality test, the Elzinga-Hogarty test and transportation cost analysis.  However, since quantitative analysis requires complete data and there are many intrinsic or extrinsic factors that may result in biased or erroneous estimation results, if relevant numbers and data are not complete in its application to a specific case, the facts associated with the individual case, the evidence under investigation and the materials obtained as a result of the investigation are still used to define the relevant market.

In practice, since it is difficult to obtain the calculation data required for quantitative analysis, qualitative analysis using observations and interviews to present the substitutability between products appears to be quite important.  Assertions raised by enterprises which file their merger in connection with the definition of the relevant market mostly define the relevant market based on qualitative analysis.  The Fair Trade Commission (hereinafter, the “FTC”) also uses qualitative analysis to define markets in most cases.

Quantitative analysis definitely required for defining the relevant markets?

Point 6 of the Principles of the Fair Trade Commission for Handling the Definition of the Relevant Market provides: “When defining the relevant markets, this Commission will consider if there is any reasonable substitutability between relevant goods or services and geographic areas and other goods or services and geographic areas; and the Commission may use the cross elasticity test and the hypothetical monopoly test to define the relevant markets.  However, the handling of individual cases is not limited to the above analyses and tests, and there is no priority sequence for the use of all kinds of analysis and test. ”

As to whether quantitative analysis is required for the relevant market, the FTC’s Frequently Asked Questions and Answers concerning the definition of the relevant market indicate: “Although quantitative analysis for the relevant market is more persuasive, still if relevant data are incomplete in its application to a specific case, a factual analysis based on the impact of the action in question on market competition activities may still be conducted to properly define the market according to the facts associated with the case, evidence under investigation and the data obtained as a result of the investigation.”

Since many factors should be considered for individual cases, different analytical methods may have different results for identical cases.  Therefore, it is necessary not only to consider the nature and types of goods or services provided by an enterprise in its business activities in the determination of specific cases but also to take into account the impact of the behavior involved in a specific case on competition to define the market.  In addition, the emphasis should be put back on the nature of the case by considering factors such as the reasons of disputes, characteristics of the products and industry, attributes of the data, etc. and the method for defining the market should be prudently selected to properly define the relevant market.

Empirical investigation required for defining the relevant market? 

In a case involving freshly brewed coffee in convenience stores, the administrative court criticized the FTC for its failure to conduct the SSNIP test and to conduct a questionnaire-based consumer survey, not to mention that the other qualitative factors proposed by the FTC did not use data obtained from participatory observations and in-depth interviews.  Therefore, the administrative court ruled that the FTC’s practice of making a determination merely on the basis of written materials without actual market evidence is not sufficient to define the relevant market of this case.  It is noteworthy that the administrative court did not criticize the FTC for making an unlawful determination due to its failure to conduct quantitative analysis in such case.  Instead, the administrative court criticized the FTC for making a rash determination due to its failure to analyze data collected from in-depth observations and interviews and to its determination of the market without sufficient concrete evidence (see the 102-Su-Geng-One-Zi-54 Decision of the Taipei High Administrative Court and the 103-Pan-Zi-195 Decision of the Supreme Administrative Court).

Although the administrative court did criticize the FTC for its definition method involving no empirical investigation of the market in the case of freshly brewed coffee in convenience stores, still this does not indicate that administrative courts have tilted towards market empirical investigation as requirement for defining a market.  With respect to the question of whether empirical investigation is definitely required for defining the relevant market, the administrative court explained in the merger of Cashbox and Holiday: “Pursuant to the 103-Pan-Zi-195 Decision of the Supreme Administrative Court, …this does not mean that empirical investigation is absolutely required of the defendant before any market is defined or the defendant will be found in violation of the law.  The plaintiff’s assertion that the defendant broke the law for failing to conduct a market empirical investigation before the product market for this matter was defined based on part of the above Supreme Administrative Court decision is also unacceptable” (see the 103-Su-Zi-1700 Decision of the Taipei High Administrative Court and the 105-Pan-Zi-428 Decision of the Supreme Administrative Court).  The administrative court reiterated the above gist in the case involving the failure to file the merger of cable television operators for approval (the 105-Su-Zi-412 Decision of the Taipei High Administrative Court).  This shows, according to the current practical opinions, that it is not true that when defining a market, the FTC is definitely required to conduct empirical investigation before it can define a market.  In practice, the FTC does not always conduct empirical investigation involving in-depth observations and interviews such as questionnaire-based surveys on the consumers, and quite a few of the resultant administrative dispositions are upheld by administrative courts.

In recent years, it seems that there is a trend that administrative courts gradually require the empirical tests.  For example, in a concerted action case involving 9 independent power plants (IPP) refusing to adjust the feed-in-tariffs (see 107-Pan-Zi-507 Decision of the Supreme Administrative Court), the administrative court believed that in order to define the market in this matter, Taipower should be asked, with respect to the needed electricity, to ascertain relevant facts concerning the 9 IPPs such as the circumstances and convenience (whether it is easy to switch) of choosing or switching different areas, the extent of price changes, and substitution relations of product areas.  In addition, Taipower should be requested to provide actual data (in order to consider relevant factors for the geographic market) and a table summarizing the data about its purchase of electricity during 2009 through 2014, as well as information about Taipower’s actual consideration factor in purchasing electricity from each IPP and the energy tariff rates as an important or unimportant consideration factor.  In addition, whether there is any horizontal competition over the electricity sold in non-guaranteed time slots by each IPP should also be considered, and Taipower should be asked to list the actual factors it actually considered with respect to the results shown in the table (consideration factors for the product market).  The administrative court actually implied in such case that empirical market investigation was required to define the relevant market.  The administrative court has changed by requiring the FTC to provide more empirical evidence to substantiate the reasonableness of its market definition.  For relevant enterprises in the case or other enterprises in the so-called relevant market, this allows them to provide their arguments concerning the definition of the relevant market and also facilitates the objective examination of an FTC disposition by the administrative court.  The trend for requiring empirical investigation and evidence will surely affect the practices of the FTC, which used to define markets mostly based on written and regulatory materials such as the size of enterprises, market position, prices and marketing strategies, so that the FTC will use investigation and empirical materials and data, which can better reflect the reality of the industry concerned, to define the relevant market of the enterprises concerned.

Although administrative courts have gradually required the use of diverse economic analyses to define markets in recent years, still when an enterprise is dissatisfied with the determination of a market defined in a disposition rendered in accordance with the Fair Trade Law, the current practice still does not allow the citation of the FTC’s failure to conduct empirical market investigation as the sole basis for the finding the illegality of the FTC’s disposition.  Under the current legal regime and practices, it is still necessary to revert to the consideration of whether the market definition or determination is based on erroneous facts or runs counter to the custom of the industry, market information, market trends and the reality according the relevant methods and consideration factors for defining or determining the relevant market.  In case the finding on the definition of the relevant market is to be challenged by empirical market investigation and quantitative analysis, if complete, reliable and sufficient price and quantity data and analyses such as industry data and market information which may be used for quantitative and qualitative analyses can be provided, it is more likely to win the support and acceptance of the administrative court.