Ink in cartridges for printers is often called ‘black gold’, or qualified as the ‘most expensive liquid in the world’. Manufacturers of printers sell their ink cartridges at (relatively) high prices, whereas they offer their printers for (relatively) low prices. The ‘cheap-appliance-expensive-consumable’-business model is used widely: coffee machines and pods, consoles and games, cars and spare-parts, etc. As a consequence of the relatively high prices on the aftermarket, independent suppliers try to enter such a lucrative aftermarket by offering generic products which are compatible with the machinery offered on the up-stream market. Not surprisingly, this leads to conflicts between those independents and manufacturers of appliances, because of the intellectual property rights over the machinery and consumables (e.g. generic producers offering coffee pads compatible with Nespresso– and Senseo-coffee machines and contesting the IP-rights in question, or – in the alternative – claiming that the refusal to license the IP-right is an abuse of a dominant position) and associated litigation (e.g. the Toshiba/Katun-case over advertisement of generic consumables which referred to the brand of the machinery).
In the EFIM-case, producers of generic ink cartridges (independent suppliers) – associated in the European Federation of Ink and Ink Cartridge Manufacturers (EFIM) – complained to the Commission, mainly because they were denied access to the intellectual property rights by the four, so-called ‘original equipment manufacturers’ (OEMs) of printers: Hewlett‑Packard, Lexmark, Canon and Epson. Without access to those intellectual property rights, producers of generic ink cartridges argued that they could not effectively compete with the OEMs on the (after)market for ink cartridges. EFIM considered that behaviour to foreclose the market for ink cartridges and therefore an abuse of a dominant position, which is prohibited under Art. 102 TFEU.
On 20 May 2009, the Commission decided to dismiss the complaint of EFIM on two interrelated grounds. It concluded
- that it was not likely that the OEMs held a dominant position on the secondary market of ink cartridges, since the OEMs did not have a dominant position on the primary market of inkjet printers; and
- therefore, it considered a lack of community (now: Union) interest to pursue an investigation.
The Commission relied on its aftermarket-analysis in its earlier Pelikan/Kyocera– and Info-Lab/Ricoh-decisions on similar complaints regarding the market for toners for printers, respectively toners for photocopiers. The Commission summarised its findings in the Pelikan/Kyocera-decision in its XXVth Report on Competition Policy (1995) (cf. p. 41) as follows.
(…) [P]urchasers were well informed about the price charged for consumables and appeared to take this into account in their decision to buy a printer. “Total cost per page” was one of the criteria most commonly used by customers when choosing a printer. This was due to the fact that life-cycle costs of consumables (mainly toner cartridges) represented a very high proportion of the value of a printer. Therefore, if the prices of consumables of a particular brand were raised, consumers would have a strong incentive to buy another printer brand. In addition, there was no evidence of possibilities for price discrimination between “old”/captive and new customers.
According to the Commission there are four criteria in order to assess the close interrelation between the aftermarket of consumables (secondary market) and up-stream market for machinery (primary market).
Dominance on the aftermarket is excluded
to the extent that a customer
- can make an informed choice including lifecycle pricing, that he
- is likely to make such an informed choice accordingly, and that
- in case of an apparent policy of exploitation being pursued in one specific aftermarket, a sufficient number of customers would adapt their purchasing behaviour at the level of the primary market
- within a reasonable time.
Since the bahaviour of the customers on the primary market, has a disciplinary effect on the behaviour of the OEMs on the secondary market, there is no dominant position on the secondary market, regardless of the 100%-market share on the latter. According to the Commission, EFIM did not put forward evidence (the complainant has the burden of proof) which proved that the market for ink cartridges is not interrelated with the primary market for printers (EFIM-decision, par. 25). Therefore, the Commission has assumed that there was no dominant position and has concluded that there was a lack of Union interest to pursue further investigation and rejected the complaint.
EFIM sought for an annulment of the EFIM-decision, but the General Court confirmed the findings of the Commission in its judgment of 24 November 2011 in Case T‑296/09), including – interestingly enough – the substantive analysis of dominance based on the four criteria mentioned. EFIM appealed to the CJEU, which delivered its judgment (until now available in the French and German language) on 19 September 2013.
The Court of Justice of the EU confirmed the findings of the GC and the Commission and dismissed the appeal of EFIM.
The CJEU held that
36 (…) le Tribunal a vérifié que la Commission n’a pas commis d’erreur manifeste d’appréciation en appliquant, pour rejeter la plainte, les principes qu’elle avait dégagés dans deux décisions antérieures [the Pelikan/Kyocera-decision and Info-Lab/Ricoh-decision; WG].
37 Or, il ressort desdits principes que toute position dominante sur les marchés secondaires, à savoir les marchés des cartouches d’encre, peut être exclue s’il est établi qu’il existe une concurrence sur le marché primaire, à savoir le marché des imprimantes, et si les marchés primaire et secondaires sont étroitement liés, ce qui dépend de la réunion de quatre conditions (…) [emphasis added; WG].
Customer’s key role in line with United Brands and Kodak
As a preliminary remark, it must be stressed that the Commission rejected the complaint on its discretionary power not to pursue a substantive analysis due to lack of community (now: Union) interest. This discretionary power is rather large and gives the Commission the possibility to dismiss complaints, irrespective of their merits (e.g. cf. A. Sánchez Graells, ‘T-119/09: GC backs broad Commission discretion not to pursue antitrust cases in absence of “Community interest”’, European Law Blog 18 September 2012). Nevertheless, the Commission made a substantive analysis in its EFIM-decision, by letting the dominance test of the four criteria it applied in its earlier aftermarket-decisions play a role in the analysis of the lack of Union interest. This has led the General Court and ultimately the CJEU (par. 37) to explicitly approve the Commission’s analysis on dominance on aftermarkets.
The role of the customer is key in these four criteria. The customer must be well informed (not only about the price of the primary product, but also about the prices on the aftermarket and take both prices into account by making life cycle cost calculations) and timely change his behaviour on the primary market to discipline OEMs which deliver poor quality/high prices in the aftermarket. In my opinion, this key role of the customer is a logical consequence of the definition of dominance given by the CJEU in the United Brands– and Hoffmann-LaRoche-cases. A dominant undertaking is able to
behave to an appreciable extent independently of its competitors, customers and ultimately of its consumers.
(par. 65, respectively, par. 38)
Even though the OEMs hold a 100%-market share in the aftermarket, as long as they cannot act independently of their customers/consumers, they do not hold a dominant position. The four criteria are a test to identify whether an OEM can act independently of its customers in the aftermarket. If the four criteria are met, an OEM is not independent of its customers and is therefore not in a dominant position. That is in line with the definition from the United Brands-case.
Now it seems to me that, in the EFIM-case, the CJEU confirms the approach taken by the Commission and that the four criteria are a rightful method to analyse dominance in an aftermarket. I do understand this dominance-test as a two fold test: (i) a test of customer’s behaviour (i.e. making life cycle cost calculation, cf. infra); (ii) in combination with the level of competition on the primary market. In my opinion one cannot conclude from the EFIM-case that fierce competition in the primary market always leads to the conclusion that there is no dominance in the aftermarket (although one may get that impression at first sight when reading par. 37 of the CJEU’s judgment, quoted above).
When the Commission applied these four criteria for the first time in the Pelikan/Kyocera-decision, it stated in its XXVth Report on Competition Policy (1995) (cf. p. 41) that its approach is similar to the US Supreme Court’s approach in the 1992 Kodak-case; also referred to as ‘Kodak aftermarket doctrine’. Although scholars perceive an erosion of that doctrine (cf. D.A.J. Goldfine, K.M. Vorrasi, ‘The Fall of the Kodak Aftermarket Doctrine: Dying a Slow Death in the Lower Courts’, Antitrust Law Journal 2004, p. 209 and D.T. Hibner, ‘Aftermarket Monopolization Claims Dismissed as Afterthought’, Antitrust Law Blog 11 July 2011).
Customers must calculate life cycle cost
In the 2005 Commission staff discussion paper to exclusionary abuses (at p. 68 et seq.) attention was paid to aftermarkets as well. The Commission gives some more practical guidance on those four criteria.
The amount of information available to consumers is an important factor for assessing the extent to which the customers, when buying the primary product, make a calculation of the overall cost of the bundle. The information available must enable customers to make accurate calculations. This is more likely to be so when the secondary product is a consumable used with the primary product in fixed proportions, than in the case of spare parts and services. (par. 257)
Moreover, for this competitive constraint from the primary market to function effectively, a sufficient number of customers must engage in life cycle cost calculations, and the supplier concerned must not be able to discriminate between customers that make such calculations and those that do not. For instance, a primary product may be purchased by both private and professional buyers. If only professional buyers make (accurate) life cycle calculations, the supplier may still have substantial market power in the aftermarket vis-à-vis private customers. (emphasis added; WG; par. 258)
In its 2010 Guidelines on Vertical Restraints, the Commission specifies that
in the case of long-term contracts or in the case of after- markets with original equipment with a long replacement time, it becomes difficult for the customers to calculate the consequences of the tying. (par. 217)
Primary and aftermarket form one single product market
Abuse of dominance is not the only field in which aftermarkets play an important role. The Commission also pays attention to the question whether aftermarkets are interrelated to the primary market in its 2010 Guidelines on Vertical Restraints. The four-criteria test finds its way in those guidelines too:
In practice, the issue is whether a significant proportion of buyers make their choice taking into account the lifetime costs of the product. (Par. 91)
And the Commission continues that paragraph, by stating:
If so, it indicates there is one market for the original equipment and spare parts combined.
It seems to me that the Commission is of the opinion that when customers make a life cycle calculation of the bundle of machinery (bought on the primary market) and products/services (bought on the aftermarket) it does not only indicate the lack of dominance on the aftermarket, but might even point at a relevant product market which at the same time consists of both the machinery and the products and/services on the aftermarket (e.g. consumables, spare parts, maintenance/repair services).
In the 2005 Commission staff discussion paper to exclusionary abuses, the Commission takes into account another criterion to assess whether aftermarkets form one and the same market with the primary market (which the Commission denominates “bundle” or “systems” market) or whether they are independent product markets, namely switching costs.
[I]t may be possible to switch to another primary product and thus avoid the higher prices in the aftermarket. This would require that switching costs are not so high as to make such an option too expensive.
If the conclusion is reached that there is no separate aftermarket, the analysis must be conducted on the overall “systems” market. Dominance would then have to be established on this market and not on a separate aftermarket. (par. 249)
Next to the switching costs-test, the Commission considers the existence of two seperate markets when
it is possible to switch to the secondary products of other producers [e.g. white label products; WG]. (par. 248)
It seems to me that the Commission uses the four criteria-test – which mainly comes down to customers making life cycle cost calculations – two times:
- first, to determine whether the primary market and aftermarket constitute one and the same “bundle” or “systems” market or whether they are seperate/independent markets and
- secondly, to determine whether the customer’s behaviour on the primary market can have disciplinary effect on the aftermarket.
This quizzes me. In the event customers make life cycle cost calculations, which leads to the conclusion that the primary and aftermarket constitute one and the same “bundle” market, one does not need to apply the same test (again) to determine the dominance on the aftermarket, since there is no seperate aftermarket at all (safe for aftermarkets where white label products are offered).
Whatever may be of this, the disciplinary effect of customers’ behaviour on the primary market demonstrates lack of dominance of the OEMs, regardless of the primary market being independent of the aftermarket or constituting one and the same “bundle” or “systems” market.
Abuse on separate/independent aftermarkets
Only aftermarkets which meet the four-criteria-test and/or switching costs-test are interrelated with the primary market and constitute one single relevant product market with the primary market. When customers cannot exercise the needed disciplinary effect on the primary market, the markets are not interrelated and the aftermarkets form a relevant product market of their own. OEMs with IP-rights over their machines/consumables probably have a dominant position on those independent aftermarket (so called proprietary aftermarkets), regardless of their position on the primary market.
In the 2005 Commission staff discussion paper to exclusionary abuses, the Commission stated that when it comes to aftermarkets which are independent of the primary markets, the Commission
presumes that it is abusive for the dominant company to reserve the aftermarket for itself by excluding competitors from that market. Such exclusion is mostly done through either tying or a refusal to deal. The tying can come about in the various ways described in the section on tying. The refusal to deal may, for instance, involve a refusal to supply information needed to provide products or services in the aftermarket; a refusal to license intellectual property rights; or a refusal to supply spare parts needed in order to provide aftermarket services. (par. 264)
Milestone examples of abuse on aftermarkets which are considered independent of the primary market are:
- the 1979 Hugin-case on the refusal to supply spare parts for cash registers to independent maintenance/repair companies, although Art. 102 TFEU did not apply due to lack of effect on interstate trade (cf. for a comparison of this case to the US Kodak-case: W.R. Andersen, ‘Kodak and the Aftermarket Tying Analysis: some Comparative Thoughts’, Pacific Rim Law & Policy Journal 1995, p. 277);
- the 1988 Volvo/Veng– and Renault-cases on the refusal to supply car spare parts to independent repairers;
- the well known 1991 Hilti-case on the tying of nail guns and nail cartridges; and
- the 1994 Tetra Pak II-case on the tying of milk cartons to filling machinery.
The Commission’s four-criteria test is now backed by the CJEU. The original printer manufacturers did not become the 21st century Hilti or Tetra Pak, because EFIM could not prove that the customers’ behaviour did not have disciplinary effect on the OEMs. Nevertheless, when the markets can be separated, OEMs operating on both markets and locking-in customers with tying arrangements and/or locking-out competitors by refusing to deal (i.e. licensing IP-rights or supplying spare parts) can still expect the vigour of the Commission.
 Nespresso compatible pods are offered by for example the Ethical Coffee Company (ECC) and its litigation in Germany: Oberlandesgericht Düsseldorf 21 February 2013, Nestlé/Ethical Coffee Company (ECC), I-2 U 72/12 and I-2 U 73/12; and by Dualit in the UK: England and Wales High Court (Patents Court) 22 April 2013, Nestec SA & Ors v Dualit Ltd & Ors  EWHC 923 (Pat)
 Dutch and Belgian Senseo coffee pad cases. The Dutch Supreme Court ruled in favour of the independent suppliers: Hoge Raad 31 October 2003, Sara Lee/DE vs. Coöperatieve Inkoopvereniging Intergro B.A., ECLI:NL:HR:2003:AI0346; the Antwerp Court of Appeal, ruled in favour of the OEM: Hof van Beroep Antwerpen 8 November 2005, Sara Lee/DE N.V. v. N.V. Fort Koffiebranderij, S.A. Cafe Liege and N.V. Beyers Koffie. Finally the European Patent Office revoked the patent over the combination of machine and coffee pad by its decision of 30 August 2006, ECLI:EP:BA:2006:T045205.20060830.
 Such as the milestone cases IMS Health and Microsoft. And specificaly related to ink cartridges: the judgment of the Oberlandesgericht Düsseldorf, 20 January 2011, Case I-2 U 92/10; cf. N. Hölzel, ‘The German Higher Regional Court Düsseldorf specifies the requirements for granting a compulsory license under antitrust law (Zwangslizenzeinwand)‘, e-Competitions Bulletin January 2011, N° 48587.
 Cf. S. Borenstein, J.K. MacKie-Mason, J.S. Netz, ‘Exercising Market Power in Proprietary Aftermarkets’, Journal of Economics & Management Strategy 2000, p. 157.
 For an analysis (in Dutch) cf. S.C. van Loon, Licentieweigering als misbruik van machtspositie, Amsterdam: deLex, 2008, p. 207-208. Other examples of independent aftermarkets are the 1991 Varta/Bosch-decision of the Commission on a separate market for replacements of car batteries and Lucas/Varity-decision of the Commission on the automotive electronics aftermarket.