On 19 April 2021 the Competition Board published its reasoned decision regarding its fully fledged investigation into whether six fertiliser suppliers(1) violated Article 4 of Law 4054 on the Protection of Competition, which prohibits, among other things, anti-competitive agreements and concerted practices (Fertiliser Suppliers, 26 November 2020, 20-51/718-317).
This decision comprises important Competition Board assessments and findings as to how insufficient documentary evidence can be supported by economic evidence to prove the existence of an anti-competitive agreement when the market demonstrates parallel and simultaneous price increases.
This article examines the facts surrounding the case, as established by the Competition Board, and how the board used economic evidence to reach its decision.
In October 2018 the Competition Board initiated a preliminary investigation ex officio into the concerned undertakings to examine reported price increases in the fertiliser market. Within the scope of the preliminary investigation, the Competition Board conducted onsite inspections and gathered evidence that led to a fully fledged investigation in February 2019. One document in particular, referred to as 'Evidence-20', was of decisive influence in leading the case handlers to suggest that the Competition Board initiate a fully fledged investigation into the undertakings.
Evidence-20 was an internal market intelligence report by Bandirma Gübre Fabrikalari AS (BAGFAS), which was in the form of an email. The email had been forwarded from BAGFAS to EGE Sanayii AS (which was part of the same group of companies), in whose premises it had been discovered during the onsite search. In the email, it was noted that one of BAGFAS's employees had informed EGE management that Istanbul Gübre Sanayii AS (IGSAS) (a competitor under the same investigation) reportedly told Gemlik Gübre Sanayi AS (GEMLIK) (another investigated competitor) that IGSAS was to increase its prices, so that GEMLIK would, in turn, also increase its prices. However, the wording of Evidence-20 was ambiguous and did not allow the Competition Board to determine the specific source of such market intelligence or when the price increases had allegedly taken place. Thus, the Competition Board decided to collect more evidence within the scope of the fully fledged investigation in order to understand the specifics of the alleged increase or ascertain whether any increase had actually occurred.
Following the initiation of the fully fledged investigation, the case handlers conducted further dawn raids on the six undertakings. While these dawn raids revealed valuable information regarding the dynamics of the market, they did not uncover any further evidence relating to what Evidence-20 suggested (ie, that GEMLIK and IGSAS had exchanged a future price strategy). Accordingly, an extensive market test was conducted with a view to fully understanding how the fertiliser market worked. The case handlers used the detailed information that they were able to obtain regarding the characteristics of the market, combined with the data collected from the investigated undertakings through information requests regarding the actual prices realised in the market.
The Competition Board found that although there were approximately 1,200 undertakings in Turkey engaged in the production or import of fertilisers, the six which were under investigation represented 80% of the market. Therefore, the Competition Board concluded that the fertiliser market had an oligopolistic structure with a small number of major players.
The Competition Board further established that there was a high dependence on imports of raw materials and finished products. The undertakings – including the investigated suppliers, some of which also imported finished fertilisers – purchased their raw material inputs and finished products from the same foreign players. Therefore, the investigated undertakings had a similar price and cost structure:
- Considering that imports required payments to be made in a foreign currency (mostly US dollars and euros), the prices in the fertiliser market were highly dependent on exchange rates. The risks surrounding changes in the exchange rates were borne by the market players and could directly affect profitability.
- The non-exclusive dealership system in the sector had resulted in a transparent market. Dealers worked with multiple fertiliser suppliers at the same time. Consequently, dealers had significant information about the price and payment terms applied by different companies and used this information to negotiate with suppliers.
- Another aspect that made the market transparent was that fertiliser suppliers often used the same port and warehouses, or the same cargo discharge services. As a result, the suppliers could obtain valuable information regarding their competitors' supply channels and conditions, such as the country of origin, type and quantity of the product. This information was considered to affect the pricing behaviour of the undertakings.
- The seasonality of the product caused demand to significantly increase during certain periods (high seasons), which translated into seasonal fluctuations in price.
- The fact that the competitors were also regular suppliers of each other increased transparency in the market.
Based on these findings, the Competition Board established that the products were homogenous and that the market was transparent and traceable.
The Competition Board first assessed the relevant product market and established various segments for the fertiliser products. On the other hand, the Competition Board also emphasises the homogeneity of the products many times throughout the decision. Based on the decision, it can be assumed that the Competition Board referred to the products in those individual segments instead of all of the products, regardless of the segment, when it dealt with the issue of homogeneity. That is to say, the Competition Board may have established that each segment included homogenous products.
The Competition Board then looked into the undertakings' market shares in different segments of the fertiliser market. It found that although there appeared to be consistent market leaders in certain segments, the market shares of the undertakings in the relevant segment were generally unstable, which could signal a competitive market.
The Competition Board went on to compare the undertakings' prices. It re-emphasised that the prices were highly dependent on price fluctuations abroad, as well as foreign currency exchange rates. In addition, the Competition Board stressed that Gübre Fabrikalari TAS (GUBRETAS) was the maximum price setter in the market since it provided the most advantageous prices to farmers (via the Agricultural Credit Cooperative, which comprised the largest buyer group), due to the structural links between the cooperative and GUBRETAS. Those prices in turn were naturally regarded as the price cap in the market since the most significant parameter deciding competition in the market was price. In light of these findings, the Competition Board established that the costs and prices remained similar among the undertakings, at least those in the same segment.
Since Evidence-20 suggested the exchange of a future pricing strategy between GEMLIK and IGSAS, and explicitly set out the alleged future price, the Competition Board first looked into whether the alleged future price could be observed in the prices of GEMLIK and IGSAS. In doing so, comparisons were made of both the listed prices and the actual prices realised in the market. The board concluded that the alleged future price had not been realised in the market.
The Competition Board then compared the prices of GEMLIK and IGSAS and observed parallels between them. However, the Competition Board emphasised that this may have been the result of the nature of a transparent market, where the product was homogenous and the competitors' prices could easily be estimated, even tracked daily, thanks to the non-exclusive dealership system observed in the market (as explained above).
After assessing the market shares and prices, the Competition Board stated that although the wording of Evidence-20 suggested an information exchange between GEMLIK and IGSAS, the concerned evidence was in the form of a market intelligence report of a third-party competitor (ie, BAGFAS) discovered in another competitor's (ie, EGE's) premises. Therefore, it was concluded that the mere statements of third parties and the price parallels in and of themselves were insufficient to establish an information exchange between the undertakings.
Despite the already detailed assessment, the case handlers requested that the Competition Authority's Economic Analyses and Research Department (EARD) conduct further economic analyses regarding the prices. Depending on the outcome, the case handlers apparently planned to conduct further dawn raids to collect more evidence to reveal all of the facts surrounding Evidence-20 or suggest that the Competition Board establish the violation.
The Competition Board stated that some of the most notable behavioural analyses for the detection of cartels are the tests that assess a structural break. A structural break, if established, may signal a violation. The Competition Board emphasised, by referring to a certain authority,(2) that, generally speaking, the results of economic analyses cannot be regarded as decisive and final evidence, although they point to a violation.
As such, the EARD was to determine whether the pricing trends could be explained by the external and internal cost shocks on the assumption that the demand conditions were homogenous. If the shocks were insufficient to explain the trends, there may have been a violation which had yet to be revealed.
In this context, the Ordinary Least Square Based Cumulative Sum (OLS-CUSUM) test was the preferred structural break test, as it was deemed to be more compatible in cases where there is a suspicion of a cartel in a given market but the relevant authority cannot be sure as to the existence and the period in which the cartel existed. The following factors were taken as the variables of the test for this particular case:
- foreign exchange rates;
- the producer price index (for the energy);
- urea and ammonium prices (inputs for the fertiliser); and
- seasonality.
The EARD analysed the price series on a monthly basis by taking into account the past 71 months and concluded that the test did not reveal any structural breaks in the prices, which would otherwise indicate an explicit or tacit agreement between the investigated undertakings. The EARD then ran another test, known as the 'Ramsey Regression Equation Specification Error Test', with a view to confirming the accuracy of the models subjected to the OLS-CUSUM test. The EARD confirmed that the changes in the undertakings' prices could be explained with the above indicated variables and concluded that the prices did not indicate any potential violation.
As a result of the economic analyses conducted and documentary evidence gathered by the Competition Board, it was established that the investigated undertakings had not violated Law 4054.
The Competition Board's explanations as to the value of the economic evidence gathered in the investigations are of particular importance for future Competition Authority investigations. As such, the Competition Board indicated that even in instances where the authority is unable to collect documentary evidence to prove a violation, economic evidence can be used to determine whether the undertakings have potentially violated competition law. On that front, the Competition Board also referred to certain authorities that suggest that the results of an economic analysis (ie, economic evidence) generally cannot, in and of themselves, be regarded as decisive or final evidence to prove a violation. However, the results might signal a potential violation, which may require more documentary evidence.
This detailed assessment of economic evidence by the Competition Board should not come as a surprise, as its recently published Handbook of Economic Analyses Used in Turkish Competition Board Decisions(3) sets out the value and usefulness of economic tests which it conducts in investigations and merger control proceedings.
Endnotes
(1) The investigated undertakings were Toros Tarim Sanayi ve Ticaret AS, BAGFAS, EGE, GUBRETAS, GEMLIK and Istanbul Gübre Sanayii AS.
(2) The authority referred to was Crede, CJ, 2019, "A Structural Break Cartel Screen for Dating and Detecting Collusion", Review of Industrial Organization, 54:543–5574.
(3) Turkish Competition Authority Economic Analyses and Research Department (2019) Handbook of Economic Analyses Used in Turkish Competition Board Decisions.
Ekrem Kalkan, competition economics counsel, contributed to the preparation of this article.
