with Şahin Ardıyok & Fırat Eğrilmez, originally published @Mondaq –
One of the interesting antitrust cases in Turkey in 2017 is the investigation of the Turkish Competition Authority (TCA) concerning 13 international banks, for their conducts in syndicated loan markets. Even though the TCA has given its decision at the end of 2017, the full decision published recently. In this brief article, we intend to assess the interesting points in the decision from a critical perspective.
Background of the Case
In April 2016, the TCA started an investigation about 13 international banks for their certain conducts in corporate loan markets. The authority started the investigation with respect to Article 4 of the Competition Act (equivalent of Article 101 of TFEU) based upon a leniency application made by one of the undertaking under scrutiny; Bank of Tokyo- Mitsubishi UFJ Turkey A.Ş. (BTMU). The TCA claims that the mentioned banks, which supply loans to corporations, have been regularly exchanging some sensitive information -including information on prices and other related conditions- in order to restrict the competition.
Among other undertakings subject to investigation, two international banks are found to be the parties of an anti-competitive “agreement/concerted practice” with BTMU separately, on the grounds that they have carried out distinct yet similar behaviors. The TCA concludes that the agreement/concerted practice between BTMU and those banks has enabled them to exchange information concerning prices, amount, maturity, or participation.
As a result, TCA decides to impose administrative fines to those three banks. However, BTMU, which have applied to the leniency program, was fully exempted from the imposed fine.
Although there are several grounds on which the decision could be elaborated, we will be focusing on five issues from a critical perspective.
Nature of the Corporate Loans Subject to the Investigation
When the full decision is assessed, one may argue that most of the corporate loans supplied by the investigated banks are multilateral loans, which are generally referred to as syndicated loans. In fact, “syndicated loan” points out to a general concept that label multilateral loans in which two or more lenders jointly agree to provide a loan to a borrower. In that sense, not only competition but also cooperation and information sharing are required to formalize an efficient syndication. This renders the investigation more complex than a regular investigation for bilateral loan transactions.
However, in general sense there is little guidance on how competition law may be applied to multilateral loans. Although the Loan Market Association (LMA) in United Kingdom published a notice on the application of competition law to syndicated loans, it provides little guidance on how banks should undertake their commercial activities during the course of the transactions that require exchange of information and cooperation, to avoid committing an infringement of competition law.
In that sense, the TCA’s decision might be considered as a significant example revealing how the practices of banks would be treated over the course of an investigation. Unfortunately, the decision provides no further guidance than the LMA’s notice.
Relevant Market Definition
Perhaps one of the most interesting part of the TCA’s decision is its assessment on market definition. In the full decision, it is surprising that the TCA do not consider the relevant market definition as necessary to evaluate the conducts of the banks. It is surprising because, market definition is the initial and indispensable part of a competition analysis, especially for the information exchange cases. On two axes -product and geography- it draws the boundaries for further phases of the competition analysis. From a methodological perspective, market definition can be considered as an instrument to reduce the complexity of market interaction. Moreover, a proper market definition isolates the market in hand, in terms of its boundaries related with both products and geographical areas.
On the contrary, the TCA argues that the market definition is not crucial for an investigation subject to Article 4 (equivalent of Article 101 of TFEU). However, market definition is required to make a proper assessment regarding the nature of information exchange, which per the Guidelines on Horizontal Cooperation Agreements of the TCA (Guidelines) is a crucial element to determine whether information exchange is anti-competitive or not. While including its remarks on the relevant market and the characteristics of the market at stake, the TCA do not make an exact definition for the market and fails to assess the main characteristics of the market that are crucial for determining the anti-competitive nature of the information exchange, such as market transparency, degree of concentration in the market and complexity of the market. Additionally, the elements such as stagnancy of the market or similarity (symmetry) of the players in the market are not regarded as well, in contrast with the instructions on determining whether information exchange would result to a collusive outcome or not, which are framed in the Guidelines.
In fact, relevant market definition is not crucial for unproblematic merger cases. However, in case of an infringement decision, lack of a proper market definition calls forth problematic issues. For instance, the calculation of imposed fines is the immediate issue that may emerge. In the full decision, the TCA determines the amount of fines considering the investigated parties’ turnover of corporate and commercial credits loans in Turkey. From this perspective, the relevant market may implicitly be considered as “commercial and corporate loans in Turkey“. However, given that the allegations predominantly concerns multilateral corporate (syndicated) loan transactions, such definition appear to be too wide in terms of product market definition, yet it is too narrow in terms of geographical coverage.
An Enlarged Approach for the Implementation of the Leniency Program
We would like to remind that the investigation launched, in the wake of the leniency application submitted by BTMU. In its decision, TCA decides that with the two other banks, BTMU have also infringed the competition law. However, because of its active role of disclosure of the infringement, TCA grants full immunity to BTMU.
In fact, this is the first time that the TCA has granted immunity without a proper cartel decision. Although the Competition Act sets forth full or partial immunity for the undertakings that actively cooperate with the TCA in certain circumstances, secondary legislation explicitly requires the applicant of leniency to reveal a “cartel” to benefit from the immunity. More concretely, per the secondary legislation the existence of a cartel agreement between the undertakings is the main condition to be deemed immune from an administrative fine.
To overcome this issue, the TCB rules that the investigated parties have infringed the competition law through an “agreement/concerted practice“. In doing so, the TCB solves the shortcoming stemming from the secondary legislation, since cartel definition made in the mentioned legislation includes agreements and concerted practices, but not information exchange. Through this approach, we understand that in the Turkish case law, anti-competitive information exchanges may also provide a basis for the leniency program.
An Interesting Interpretation of the Effects Principle
One of the key questions for the validity of domestic competition rules is whether substantial, direct, and foreseeable effects on the domestic market would suffice to justify the application of domestic competition law. This is generally known as “the effects principle”.
In its decision, TCA adopts an interesting interpretation of the effects principle. TCA assesses that if an undertaking in Turkey has a subsidiary in another country, any infringement claim related with this subsidiary has an effect in Turkey through the main undertaking. This is interesting because TCA -without looking at effect- automatically grants authorization itself to investigate the claims all over the world through the undertakings established in Turkey.
Fruit of the Poisonous Tree
Fruit of the poisonous tree is a legal metaphor that extends the exclusionary rule to make evidence inadmissible in court if it was derived from evidence that was illegally obtained. The logic of the terminology is that if the source (the “tree”) of the evidence is tainted, then anything gained (the “fruit”) from it is tainted as well.
As it is seen in the full decision, some of the critical claims about the investigated parties were based on phone (voice) records. The defendants described those documents as a “poisonous tree” and the evidences obtained from those records as the “fruit of the poisonous tree”, pursuant to the applicable rules stipulated in the criminal and constitutional sources. However, the TCA mentions that Competition Act gives enough authorization to the TCA to use such evidences. It also insists that voice recording is a common application in the banking industry and thus the employees who are involved in the alleged conduct had consent for their conversation to be recorded. As a result, the TCA argues that the element of consent deems the usage of voice records lawful and it is legally convenient to use them as a means of proof regarding the conduct at stake, which was deemed as a violation of the Competition Act in the wake of the investigation.
 Turkish Competition Board Decision, numbered 17-39/636-276 and dated 28.11.2017,
http://www.rekabet.gov.tr/Karar?kararId=b8a26358-485b-4af7-9d42-dc40652899fb> accessed May 2, 2018.
 In the full decision the banks are listed as follows: Bank of Tokyo- Mitsubishi UFJ Turkey A.Ş. (BTMU), Citibank A.Ş. (CITI), Deutsche Bank A.Ş. (DB), HSBC Bank A.Ş. (HSBC), ING Bank A.Ş. (ING), JPMorgan Chase Bank İstanbul Branch (JP), Merrill Lynch Yatırım Bank A.Ş. (BOFA), Société Générale S.A. Paris Merkezi Fransa İstanbul Türkiye Merkez Şubesi (SG), Standard Chartered Yatırım Bankası Türk A.Ş. (SC), Sumitomo Mitsui Banking Corporation (SMBC), The Royal Bank of Scotland İstanbul Branch (RBS), Türk Ekonomi Bankası A.Ş. (TEB), and UBS AG (UBS).
 There is an existing financial economics literature how the availability of information affect the structure of syndicated loans, e.g.SUFI, A. (2007). Information Asymmetry and Financing Arrangements: Evidence from Syndicated Loans, Journal of Finance, 62(2), s.629–668.
 LMA (2014). Notice on the Application of Competition Law to Syndicated Loan Agreements, http://www.lma.eu.com/application/files/1514/6695/7414/Notice_on_the_Application_of_Competition_Law_to_Syndicated_Loan_Arrangements.pdf> accessed May 2, 2018.
 HÜSCHELRATH, K. (2009). Competition Policy Analysis: An Integrated Approach (Vol. 41). Heidelberg: Physica-Verlag HD, Ch.2.
 GEROSKI, P. A. (1998). Thinking creatively about markets, International Journal of Industrial Organization, 16(6), p.677-695.
 Guidelines on Horizontal Cooperation Agreements; <http://www.rekabet.gov.tr/Dosya/guidelines/7-pdf> accessed May 2, 2018.
 Act No. 4054 on the Protection of Competition, Article 16/d, <http://www.rekabet.gov.tr/Dosya/geneldosya/kanuning.pdf> accessed May 2, 2018.
 Regulation on Active Cooperation for Detecting Cartels (Active Cooperation/Leniency Regulation), <http://www.rekabet.gov.tr/Dosya/regulations/1-pdf> accessed May 2, 2018.
 Cornell Law School, Legal Information Institute; <https://www.law.cornell.edu/wex/fruit_of_the_poisonous_tree> accessed May 2, 2018.