Author(s): Pablo Solano Diaz
Source: European Journal of Legal Studies, Volume Issue ()
Price Parity Clauses:
Has the Commission Let Slip the Watchdogs of War?
Pablo Solano Díaz*
While the proliferation of online markets has presented competition authorities with a string of challenges of diverse nature, the price parity phenomenon depicts particularly well the Commission's 'fight-chaos-with-chaos' approach to digital economy. Indeed, the EU watchdog's passive attitude towards the multiple enquiries into most-favoured-nation clausesby different national trustbusters across the Old Continent risks going down in universal history of antitrust infamy. With the awareness that rivers of ink have already been poured over the subject, this paper adopts a brand new stance by looking on its opportunity side. In the light of the timid open-mindedness recently shown by the European Court of Justice as regards the object-effect dichotomy in Groupement des Cartes Bancairesand Maxima Latvijajudgements (which seem to steer away from Pierre Fabre's sternness), I will discuss different national solutions with a view to vindicating not only a more consistent but also an unprejudiced effects-based approach that should transport analogue EU law enforcement to the digital era.
Keywords: Degree of harm, most-favoured-nation clauses, national competition authorities, price parity clauses, restrictions by effect and restrictions by object
Only eclipsed by the crusade against Google, the generalised recourse to so‑called most‑favoured‑nation (MFN) clausesby online platforms stands out among other digital‑economy anathemas for the implications vis-à-vis EU competition policy of divergent national enforcement reactions.The danger for EU law uniformity that these divergences involve has become particularly serious as far as legal qualification has gained relevance in order to determine the assessment to which the conduct should be subject after recent judgements of the European Court of Justice (ECJ).By redefining the traditionally exorbitant notion of conducts presumed to be restrictive by their own object (without need to analyse their effects), the judges in Luxembourg have paved the way for a bold policy-driven move towards the long-awaited (and never come) 'more economic approach'. Unfortunately, the Commission appears to have given up enforcement in this uncharted domain to national competition authorities (NCAs).
This paper challenges the Commission's passivity towards price parity clauses, insofar as its wait‑and‑see approach has led to blessings and curses on the side of different NCAs, ranging from the Swedishbenevolence to the German ordo-liberal admonition. Moreover, this voyeurism of the European authority is not new nor is it endemic to price parity clauses. On the contrary, it smacks of previous twists and turns to conceal a relegation of the EU interest, already condemned by the General Court in CEAHR,which I will evoke in Section II. Furthermore, other recent developments in e‑commerce ‒ that fall outside the scope of this paper ‒have befallen the same fate (e.g. bans on online selling).
To provide an insight into the issue, after justifying why EU guidelines are needed on the subject at stake, I will begin with an overview of the different theories of harm that may apply to MFN clauses and the potential efficiencies to which they may give rise. A discussion on the contrasting views taken by the Swedish, Italian, French, German and British competition watchdogs will follow in order to underscore the need for common guidance. Then, I will focus on the legal qualification of price parity clauses, which might have an impact on their consideration as either by‑object or by‑effect restrictions, now that the ECJ is revisiting this fundamental dichotomy. Finally, I will conclude by pleading in favour of a common facts‑based and effects‑oriented approach being embraced by the Commission as the best (and possibly the only) fit with digital‑world restrictions.
As a preface to the discussion that follows, it is necessary to address the question whether the Commission's shying away from leading enforcement as regards online restrictions is an unequivocally ill-conceived strategy. It is certain that the de-centralised paradigm brought about by Regulation 1/2003has borne fruits over its thirteen years of existence, but the new features of the digital economy have dramatically changed the business and economic assumptions on which its design is based. This evolution has evinced the need for some adjustments, as the Commission acknowledged (at least impliedly) by launching a public consultation on empowering the NCAs to be more effective enforcers.Of course, it does not mean that the de‑centralisation process required by the principle of subsidiarity has to be reverted, but the appropriate dose of EU action must be inoculated in order not to err on the side of excessive divergence. After all, the whole system devised by Regulation 1/2003 rests in a fair (and delicate) trade‑off between subsidiarity and proportionality, on the one hand, and uniformity and legal certainty,on the other.
Indeed, the Commission appears to be awakening to this reality, although its latest reaction -in the form of a Communication supplementing the two 2015 legislative proposals on the supply of digital content and on online and other distance sales of goods -is mainly focused on unilateral discrimination (i.e. geo-blocking and geo-filtering).It is true that concerns posed by geo‑blocking and geo‑filtering may seem to have more evident implications vis‑à‑vis the single market imperative, thereby calling for closer attention by the EU legislator at first sight. However, as far as Article 101 infringements could equally hamper cross-border e-commerce, certain guidance by the European trustbuster cannot be put off for a number of reasons.
From an economic standpoint, it is certain that different national regulatory treatments of online restrictions run counter the goals of the Digital Single Market, namely better access for consumers and businesses to goods and services offered online across Europe; greater trust for consumers and certainty for businesses with clear, less fragmented rules for online sales of goods; and lower transaction costs and administrative burden for businesses when trading online across borders.In the same vein, concerning online platforms, an EU policy approach is necessary to 'avoid fragmentation and obstacles in the Digital Single Market', insofar as 'different national rules can otherwise create uncertainty for economic operators, make scaling-up more difficult for startups and limit the availability of digital services'.
A further argument is provided by fiscal federalism, which assumes that the allocation of functions in a multilevel polity should pursue the minimisation of inter-jurisdictional spill-over effects in the provision of public goods.The assignment decision ultimately boils down to a trade-off between the losses derived from failure to internalise inter-jurisdictional spill-over effects and a second element traditionally associated with the failure to identify local preferences and linked by 'second-generation' theories to various centralisation inefficiencies (e.g. misallocations resulting from imperfect information or reduced accountability at central level).Therefore, uniform enforcement across the EU is needed to internalise cross‑border externalities inherent to genuinely supranational digital trends. Additionally, assuming that an efficient level of public good output is set by an authority whose jurisdictional scope encompasses the geographic range of the benefits arising from that public good,the Commission should decide (or at least make uniform) the efficient intensity of an enforcement action from which the single market as a whole would derive advantage.
The same reasoning, construed in terms of subsidiarity, militates in favour of the greater effectiveness of EU‑level action in an inherently cross‑border field as is e‑commerce. Subsidiarity and proportionality,enshrined in paragraphs 3 and 4 of Article 5 of the Treaty on the European Union, along with uniformity and legal certainty,preside over the Regulation 1/2003 setting, be it because the national enforcement of EU competition law can be considered a shared competenceor simply because 'the institutional arrangements governing competition are specifically designed to be in accordance with the principle of subsidiarity'.Consequently, as online phenomena by their own nature go beyond the sphere of national action, the objectives of an enforcement action aiming at them could not be in principle sufficiently achieved by NCAs, and the Commission would be better placed by reason of the action's scale or effects. Nevertheless, under the principle of proportionality, EU action cannot exceed what is necessary 'to allow the Community competition rules to be applied effectively'.
The interplay between those principles relies on the cooperation system provided for in article 11 of Regulation 1/2003. Nonetheless, in the case of MFN clauses, coordination within the European Competition Network (ECN) has led to divergent outcomes in the absence of greater involvement of the EU trustbuster. Consequently, as a first step, the Commission should issue clear guidelines on how to deal with online restrictions by means of soft‑law instruments that provide NCAs with a systematic and policy-driven theoretical foundation on which to base constructive dialogue within the ECN. Additionally, the Commission's proactive intervention as amicuscuriae in national judicial proceedings under Article 15 of Regulation 1/2003could contribute to steer the debate towards the enforcement approach proposed in Sections III.4and IV.
A more drastic alternative to be pondered over at this stage would be relieving NCAs of their EU competition law enforcement competence under article 11(6) of Regulation 1/2003, provided that the Commission considers the scenario in which the MFN phenomenon produces its effects -featured by a number of Member States being affected by an essentially cross‑border conduct carried out by EU-based actors (e.g. Booking.com), as well as non-EU based players (e.g. Expedia) -similar to the one depicted by the General Court in CEAHR:
The practice complained of exists in at least five Member States, or possibly in all the Member States, and is attributable to undertakings which have their head offices and places of production outside of the European Union, which suggests that action at European Union level could be more effective than various actions at national level.
Finally, from the policy viewpoint, if the Commission were not to lead the way into the unknown domain of online restrictions, a unique opportunity to embrace an effects‑based approach ‒ in a field particularly well suited for (and requiring) factual case‑by‑case analysis ‒would be foregone. Certainly, when it comes to digital economy, anticompetitive and procompetitive effects are not as univocal as they are in the analogue world and, thus, the room for presumptions is more limited. It is equally true that an unknown issue could benefit from national experimentation under ECN coordination, but only sound EU guidance can ward off the increasing temptation for NCAs to adopt a comfortable 'by‑object' approach to online restrictions, against which the outgoing chief economist of the Commission has recently warned.
After justifying the need for an EU approach to online restrictions, I will introduce price parity clauses as a type of restraint that illustrates particularly well the lack of a common frame of reference in the digital economy. At a second stage, I will present the divergent outcomes resulting from the Commission's failure to keep national watchdogs on a leash and discuss the importance of legal qualification as an additional factor calling for the EU common enforcement approach sketched as an epilogue to this Section.
A price parity or MFN clause can be broadly defined as 'an agreement whereby a seller agrees that a buyer will benefit from terms that are at least as favourable as those offered by the seller to any other buyer'.In practice, price parity clauses have been extensively used by multi-sided online platforms (which operate as interfaces between different groups of users whose demand is interdependent). The consolidation of business models based on multi-sided platforms gives price parity clauses the potential to affect millions of consumers purchasing all types of goods or services over the internet.
For instance, Booking.com ‒ an online travel agency (OTA) rendering free searching, comparison and booking services for users while charging hotels a proportional fee per reservation ‒imposed on hotels the obligation for the latter to offer through Booking.com equal or more advantageous prices, as well as at least equal availability of overnight stays, for the same type of accommodation, date, type of bed and number of customers than those offered through the hotels' direct channel or through competing OTAs.
Price parity clauses are only tangentially touched upon in the Guidelines on Vertical Restraints as a 'supportive' measure to make direct or indirect price fixing more effective or to make maximum or recommended prices work as fixed ones.However, since this paper vindicates clearer (or at least any) guidance on MFN clauses, I will embark upon a 'lex ferenda' exercise in this Section with a view to identifying the factors to be taken into account when analysing such clauses form an EU competition law viewpoint.
Although the effects of price parity clauses also depend on other factorssuch as likelihood of application and monitoring, position of the parties to the contract in which they are included, or market environment (i.e. degree of rivalry at the different levels of the supply chain, density of the MFN weave or market transparency), the three criteria presented below are particularly useful to classify such clauses and to identify the associated theory of harm.In any event, a case‑by‑case analysis is required.
First and foremost, the level of the supply chain at which the clauses operate has to be considered. MFN clauses may be set at the retail level, stipulating that the price and other commercial conditions applied by a retailer to the goods or services of a particular supplier cannot be less favourable than those offered by a competing retailer in relation to the same supplier's goods or services. They can also be in place at the wholesale level, establishing that the conditions offered by a wholesaler to a retailer must be at least equally favourable as those offered to a competing retailer.
Secondly, the business model is to be factored in. Whereas the wholesale model is featured by an agreement on the price that the upstream seller charges the online platform (which operates as a retailer and remains free to quote any final price), the agency model entails that the online platform acts as a agent for the seller by marketing the principal's goods of services in exchange for a commission for each sale made.In this second scenario, the seller sets the retail price without falling foul of Article 101 only to the extent that the online platform qualifies as a genuine agent in the sense of the Guidelines on Vertical Restraints.
The agency model increases the potential for restriction of parity clauses byeliminating the possibility for the principal to use the quotation of lower prices on the platform as a reward for the agent lowering its commission. However, in this setting, MFN clauses may promote competitive pricing on the platform by preventing the seller from charging a 'monopoly price' (i.e. price over the competitive level).Thus, a sufficient degree of inter‑brand competition (i.e. among sellers) could make up for the loss of intra‑brand competition (i.e. among platforms), as explored further below.
On the contrary, under the wholesale model, price parity clauses do not raise concerns as long as they do not impair the retailer's freedom to set the price.In fact, they contribute to guaranteeing a competitive cost structure for the platform.Nonetheless, by way of example, they might be considered restrictive if they have the effect of reducing the retailer's bargaining power vis‑à‑vis the wholesaler, which would encroach on the retailer's freedom to set final prices thereby precluding it from obtaining price reductions that could otherwise be passed onto consumers.
Thirdly, the scope of the clause is particularly decisive in ascertaining its anticompetitive effect. Broad MFN clauses (i.e. agreements by virtue of which a seller commits to advertising on an online platform equal or lower prices than listed on competing onlinemarketplaces) present a higher potential for reduction of price competition. In particular, online platforms protected by broad retail MFN clauses under the agency model are incentivised to charge sellers higher (or, at least, not lower) commissions since these higher commissions will not translate into higher advertised prices. Likewise, there are no incentives for platforms to reduce commissions charged on the principal as this reduction will not feed back into lower prices that would otherwise allow them to enhance their market footprint.
Against this backdrop, the first main theory of harm associated with broad retail MFN clauses is reduced intra‑brand rivalry and price uniformity across platforms. This is due to the fact that incentives for platforms to be more cost‑efficient and to pass cost savings onto consumers via lower advertised prices are undermined and barriers to entry and expansion are reinforced (since new entrants would be barred from leading a low‑price strategy). In this connection, entry foreclosure risks being greater where the entrant attempts to adopt a different business model, and where non‑price competition does not make economic sense (e.g. due to network effects that may prevent entry by more efficient platforms than the incumbents).Nevertheless, if the new entrant adopts a business model similar to the incumbents' price parity clauses, it may afford the former certain protection.Ultimately, reduced price competition amongst online platforms translates into higher prices for consumers, insofar as the seller passes higher commissions through in the form of higher listed prices, and the process continues until the seller would be better off de‑listing its product from the online platform.
Secondly, broad MFN clauses may reduce inter‑brand competition, insofar as a sufficiently dense network of broad MFN clauses discourages sellers from competing vigorously, while, at the same time, collusion among them is easier (as their ability to enforce horizontal agreements is enhanced by the limited price variety).This price effect 'similar to direct collusion'is even more worrisome to trustbusters, as shown by the Commission'sE‑Books caseor the Bundeskartellamt's HRS decision,the reason being that a sufficient degree of inter‑brandcompetitioncould compensate for the lack of intra‑brand rivalry.A contrario, even uniform prices are likely to be competitive if enough inter-brand competitionis guaranteed, which in turn depends on the number and size of platforms benefiting from parity clauses,the number of sellers bound by them,the relative bargaining power of platforms and sellers,orthe availability of other price comparison tools (i.e. metasearch engines).
Other admittedly weaker theories of harm are the hindrance of incentives for innovation and investments by their lack of impact on lower final prices, and the conveyance by broad MFN clauses of a credible indication that prices advertised on the platform benefitting from such clauses are the lowest.This second effect is only problematic if buyers are aware of the parity agreements and if they are price-elastic, as well as if other platforms do not lead the same low‑cost strategy.
Besides, narrow MFN clauses are potentially less restrictive in that the seller is only bound not to offer through its own online (and sometimes also offline) channel lower prices than listed on theplatform'swebsite. Hence, despite the existence of such clauses, the seller may still leverage on its ability to lower prices in order to push commissions down. Nonetheless, a sufficiently thick lattice of narrow parity clauses might also lead to anticompetitive restrictions. For instance, it may incentivise the principal to spread out any increase in the commission charged by the agent among the prices advertised through other platforms in order not to reduce the competitiveness of its own channel. Thus, the effect of higher commissions on the advertised price would decrease and intra‑brand competition on commissions would be stifled. Moreover, narrow MFN clauses weaken the competitive constraint imposed by the seller's direct channel on the online marketplaces, although this appears unlikely to be a problem so long as inter‑brand competition is fostered by the easing of price comparison.
Wholesale MFN clause between wholesaler and retailer 1: w1 ≤ w2
Retail MFN clause between wholesaler and retailer 1: r1 ≤ r2
Wholesale MFN clause between wholesaler and retailer 1: c1 ≥ c2
Narrow retail MFN clause between wholesaler and retailer 1: P1 ≤ P3
Broad retail MFN clause between wholesaler and retailer 1: P1 ≤ P2 and P3
Figure 1: Main types of MFN clauses according to the supply chain level, business model and scope.
Price parity clauses also give rise to remarkable benefits. First of all, they are primarily devised to solve the hold-up issue in vertical relationsby preventing the seller from free-riding on the platform's specific demand‑enhancing investments in the promotion of the seller's goods or services and brand image (e.g. marketing, advertising, after-sale services or guarantees), which also enhance the attractiveness of the online marketplace vis-à-vis other sale channels.For example, the pay-per-reservation model may incentivise hotels to use the platform to attract the customer and complete the booking through its own webpage.When this parasitism is horizontal, it can only be prevented by means of broad MFN clauses, whilstnarrow parity clauses suffice to avoid vertical free-riding.
Other efficiencies resulting from parity clauses are the avoidance of delays in transactions because of the transparency of alternative bargains and the reduction in transaction costs.For instance, when negotiating long-term contracts, buyers can be certain that sellers will not place them at a competitive disadvantage while still bound by the contract by offering better conditions to competitors later on.This, in turn, directly benefits consumers by indicating the lowest prices and by easing switching. The enhancement of other forms of competition (e.g. quality, post-sale services or advertising) brought about by the prevention of the free‑riding problem is equally noteworthy.
The Commission's wait-and-see strategy has prompted a clash of paradigms as regards MFN clauses. The mirage of the joint (and balanced) position reached by the Swedish, French and Italian NCAs in their respective Booking.com decisions,under the aegis of the ECN, has been overtly challenged by the German Federal Cartel Office.
As illustrated by the French example, the concerns over broad MFN clauses under the ECN orthodox approach were as follows: (i) the lessening of competition between Booking.com and the other distribution channels as far as hotels were deprived of their two main bargaining levers to respond to the level of commissions charged by Booking.com (i.e. retail prices and number of overnight stays);(ii) the risk of foreclosure of rival platforms, which were prevented from offering hotels lower commissions and, consequently, from passing the savings onto internet users in the form of lower prices and more availability of overnight stays;and (iii) the cumulative effect of the MFN lattice in place, which exacerbated the potential for restriction by means of increasing the market power of the otherwise atomised ensemble of OTAs.
The theories of harm listed above led to accepting Booking.com's commitments to, among others, (i) removing any parity clause regarding price and conditions vis-à-vis other OTAs and hotels' offline channel; (ii) completely removing any parity clause concerning availability; and (iii) respecting the possibility for hotels to revert to previous customers.Therefore, narrow MFN clauses affecting hotels' online channel were not found to be a problem in principle.
Particularly appealing and minimalist (albeit maybe overly simplistic) is the Scandinavian‑design approach adopted by the Swedish authority, which appears to link vertical concerns to narrow parity, on the one hand, and horizontal restrictions to broad MFN clauses, on the other. This double dichotomy led the Konkurrentsverket to conclude that horizontally problematic wide party clauses should be considered restrictive in that Booking.com had less incentives to compete by offering hotels low commission rates, thereby pushing hotel room prices up.On the contrary, narrow parity clauses are purely vertical (insofar as hotels' direct channel and OTAs do not compete on the same relevant market)and should consequently be blessed to the extent that they do not give rise to restrictions on competition on any neighbouring market. Moreover, narrow MFN clauses foster inter‑brand competition via enhanced transparency and reduce the risk that hotels free‑ride on investments made by the platform.
This lax stance of the ECN champions towards narrow price parity clauses -later confirmed by Italian and Swedish NCAs' closure of proceedings against Expedia-stands in stark contrast with the German watchdog's position. In this regard, the Düsseldorf High Regional Court's upholding of the Bundeskartellamt's decision to completely ban (broad) MFN clauses as vertical restrictions was interpreted by the latter as a full-fledged 'principle in relation to restrictions of competition in the Internet' (against which the commitments offered by Booking.com would be insufficient).Later on, the Federal Cartel Office confirmed its stern position by rejecting ECN‑designed Booking.com's commitments on the grounds that narrow price parity clauses 'also restrict both competition between existing portals and competition between the hotels themselves' (in the words of the authority's president Andreas Mundt).The order to remove price parity clauses was subsequently challenged by Booking.com and upheld by the Düsseldorf High Regional Court.
Nonetheless, a compromise remains possible so far as the Bundeskartellamt's decisions refer to the specific circumstances of the German marketand do not conclusively state the hard‑core (by‑object) nature of price parity schemes (although this has been openly suggested by Andreas Mundt). In HRS, the German watchdog highlighted that the platform's market share exceeded 30%,thereby Regulation 330/2010not being applicable regardless of the question whether the MFN clauses at hand amounted to a blacklisted hard‑core restriction.Other case‑specific circumstances were taken into consideration, e.g. active monitoring and enforcement by HRS and the particular conditions of the market, namely the existence of MFN clauses in agreements between hotels and HRS's two closest rivals (Booking.com and Expedia).This market‑specific argument has been reversed by the Italian NCA to drop the probe into Expedia after this OTA accepted to adjust its terms and conditions along the lines of Booking.com's commitments. The Italian trustbuster concluded that market circumstances (that originally posed competition concerns) changed as a result of both major players abandoning the problematic clauses and, thus, the investigation was left without purpose.
Moreover, ECN‑led Booking.com and German HRS decisions seemed to agree on the vertical nature of narrow parity,as well as on the fact that OTAs and hotels' direct channels are not active in the same relevant market,which entails that restrictions on the prices offered through hotels' own sites must not be in principle anticompetitive. This fact ‒ along with the attention paid to market specificities not only by the Bundeskartellamt but also by the Italian and Swedish NCAs in their more recent Expedia decisions ‒appears to provide the Commission with sound arguments to steer towards a common ground between these seemingly opposed paradigms. This task must be orchestrated in Berlaymont for the legal, institutional, economic and policy reasons presented in Section IIand given the inability of national watchdogs to reach the uniform response that these reasons demand.
Across the Channel, the limited concern over narrow parity clauses shown by the British Competition and Markets Authority (evinced in the closure of its enquiry into hotels' online discounting restrictions on administrative priority grounds) focused on the vertical aspect.Nonetheless, one year before, the British trustbuster had highlighted the intra-brand concerns over broad MFN clauses in contracts between private motor insurance providers and price comparison websites (PCWs), since the detriment to consumers 'was likely to be significant as wide MFNs effectively prevented price competition between PCWs'.
Points in common between ECN and Federal Cartel Office
Rifts between ECN and Federal Cartel Office
Federal Cartel Office
Ø OTAs and hotels' direct channel belong to different relevant markets.
Ø Broad price parity clauses are prohibited vertical restrictions, although not conclusively classified as hard-core (by-object) restrictions ineligible for block exemption under Regulation 330/2010.
Ø Market-specific circumstances and factual considerations may play a relevant role.
Ø Narrow price parity clauses are in principle not problematic.
Ø Commitments offered by Booking.com remove concerns.
Ø Narrow price parity clauses may restrict competition.
Ø Commitments offered by Booking.com do not sufficiently address concerns.
Figure 2: Main points in common and rifts between the ECN and German positions.
Although the ECJ rulings in Inteland Post Danmark IIappeared to have smothered the embers of the 'more economic approach' (at least when it comes to Article 102), Groupement des Cartes Bancaires and Maxima Latvija judgements recently blew on them as regards Article 101. Since discussing these rulings is beyond the scope of this paper, suffice it to say that the Court, albeit cautiously, revisited the expansionist notion of restrictions by object and placed on authorities the charge of conducting a preliminary factual analysis in order to establish that a certain degree of harm is inherent to the conduct at stake for it to be considered a restriction by object.
If this jurisprudential development were to have a resonance in the field of price parity clauses, it is contended in this paper that the preliminary assessment of the degree of harm of MFN clauses should look mainly at the factors presented in Section III.1: (i) business model and level of the supply chain at which the clause operates; (ii) scope of the parity clauses; (iii) relative market power of sellers and platforms; (iv) concentration of the upstream and downstream markets; (v) number of platforms benefitting from, and sellers bound by, such clauses; (vi) the proportion of the seller's overall sales made through the platform; (vii) or the availability of other price comparison mechanisms (i.e. metasearch sites).
Furthermore, legal qualification of MFN clauses as horizontal or vertical restrictions becomes of the utmost importance. Firstly, vertical restraints are more likely to be subject to a by‑effect exam pursuant to the recent narrower interpretation of infringements by‑object. In this vein, the Maxima Latvija judgement (which ruled out the restrictive object of a commercial lease agreement conferring on the tenant a right to veto the leasing of other premises in the same shopping centre) contrasts with the traditionalby‑object analysis carried out in the Eturas ruling(in which the vertically related platform just played the role of a hub in an alleged horizontal concerted practice).
Secondly, if concerns posed by MFN clauses were limited to the vertical plane, they would be eligible for exemption under Regulation 330/2010, with market definition coming to the forefront in that case. As hard‑core restrictions cannot be exempted under Regulation 330/2010, the by‑object issue comes full circle. In other words, not being blacklisted in Article 4 of Regulation 330/2010 as hard‑core restrictions, there is no reason to consider vertical price parity schemes prima facie excluded from block exemption, not even from the de minimis safe harbourto the extent that they are not considered restrictions by object.This would be consistent with the Commission's Guidelines on Vertical Restraints not considering MFN clauses as necessarily leading to retail price maintenance.The fact that in the Booking.com cases cited above the market share above 30% prevented eligibility for block exemptionand the case-by-case approach required by the breadth of the phenomenon at hand militate in favour of the applicability to purely vertical MFN clauses of both the Regulation 330/2010 exemption and the de minimis rule.
This proposed distinction between MFN clauses in horizontal agreements ‒ in which a by‑object analysis is more deeply engrained ‒and vertical price parity ‒ more prone to by‑effect assessment ‒mirrors to a certain extent the position taken by the American antitrust authorities. In the United States, the assessment of MFN clauses is usually conducted under the rule of reason (by pondering over direct or indirect proof of likely damages and benefits arising from the clauses) where they are not part of an agreement between competitors.It seems clear that the rule of reason as such has no place in the EU competition legal frameworkand the balancing of anti-competitive and pro-competitive effects must be conducted exclusively within the framework of Article 101(3).However, the element of appreciability has a role to play in by‑effect cases (unlike in by‑object settings),in which an appreciable anti-competitive effect cannot be presumed and Article 101(1) does not apply if the likely negative effect is insignificant.
By way of corollary, the Commission should make clear that a by‑object analysis is only appropriate where the preliminary assessment of the degree of harm reveals that the broad MFN clauses in place are a device for either horizontal coordination amongst sellers (asin the E‑Books case) or vertical price‑fixing at the level of platforms. As explained in Section III.3, the preliminary degree‑of‑harm appraisal of the economic and legal context should focus on, inter alia, the business model, the supply chain level and scope of the clause, the relative market power of sellers and platforms, the concentration of the upstream and downstream markets, the density of the MFN lattice, the proportion of the seller's overall sales made through the platform, and the availability of other price comparison mechanisms.
Narrow MFN schemes, in turn, should be subject to a by-effect assessment in view of their limited degree of harm, except where either they are devised to conceal an Eturas‑like arrangement (tantamount to a concerted practice) or a sufficiently tight weave of narrow MFN clauses produces an effect equivalent to a more subtle hub-and-spoke setting resulting in the fixing of a minimum price.
At any rate, price parity clauses that are not intended (i.e. do not have as their object) to allow for horizontal coordination at the level of sellers -or, in the case of broad MFN clauses, at the level of platforms -should be eligible for exemption under Regulation 330/2010 or even considered of minor importance under the de minimis Notice, as asserted in Section III.3.
As an additional remark, while the Eturas case seems to hint at a return to Apple horizontal suspicions, the national enquiries to which the Commission seems to have relinquished the EU‑wide approach appear to gravitate towards vertical concerns. Therefore, one could argue that a first stage in the assessment of parity clauses should revolve around determining whether they represent a stand-alone concern or whether they are only problematic as an instrument for more serious restrictions (i.e. namely collusion or resale price maintenance). After all, the reference to MFN clauses in the Vertical Guidelines is limited to their role as indirect means of achieving price-fixing.
In a nutshell, although the patchwork of national cases provides some useful inputs (e.g. focus on vertical concerns, lax stance towards narrow parity clauses, or consideration of factual and market-specific aspects), EU guidance is required to develop a systematic and coherent methodology based on the elements summarised in this Section.
EU and national trustbusters across the Old Continent have, from the very outset, showed a tendency to shoehorn any market development in the stale original legal framework come hell or high waters. This inveterate propensity has never been so worrisome as it is now in the current landscape of skyrocketing technological development. The reason is that the digital world has brought about a number of new phenomena transcending the physical borders of Member States (as prominently evinced by the MFN cases HRS, Booking.com and Expedia). Therefore, the ever‑postponed innovative response required to capture an increasingly complex reality cannot be left to unguided national discretion for the multiple reasons stated in Section II.
Furthermore, the stage seems to have been set by the Court for the Commission to cut the Gordian knot of the worn out object‑based supremacy; something that can be done without necessarily breaking away from the fundamental object‑effect dichotomy, but using existing legal categories, as explored in Section III.3. Hence, it is time for the Guardian of the Treaties to grasp the nettle and avail of the momentum created by the ECJ case‑law to lay the foundations for a new and sound effects‑oriented position towards MFN clauses and other genuinely online restrictions.
Indeed, as explained in detail in Section II, the Commission is the best placed to develop a systematic and coherent enforcement approach based on the elements listed in Section III.4, which NCAs have failed to produce so far. Setting forth clear guidelines on e‑commerce (that NCAs could subsequently apply in national proceedings) would contribute to the Digital Single Market by creating a more business‑friendly environment (that refrains from strangling new business models) and by safeguarding legal certainty.
We live in a brave new world, as Shakespeare poetically remarked more than four centuries ago, that calls for an equally not only brave, but also new response by the whilom motor of European integration that once again lags behind long-legged (but often short‑sighted) national watchdogs.
* LL.M. (European Law and Economic Analysis, College of Europe, Bruges, Belgium) and EU and competitionassociate at a leading Spanish law firm, e-mail: firstname.lastname@example.org.
 Cases C-67/13 P, ECLI:EU:C:2014:2204, Groupement des Cartes Bancaires, and C-345/14, ECLI:EU:C:2015:784, Maxima Latvija.
 Case T-427/08, ECLI:EU:T:2010:517,CEAHR v Commission, paras 157-178.
 Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty  OJ L1/1.
 Commission, 'Empowering the national competition authorities to be more effective enforcers' <http://ec.europa.eu/competition/consultations/2015_effective_enforcers/index_en.html> accessed on 10 June 2016.
 Commission, 'Commission updates EU audiovisual rules and presents targeted approach to online platforms' (Press release, 25 May 2016) <http://europa.eu/
rapid/press-release_IP-16-1873_en.htm> accessed on 10 June 2016.
 Commission, 'Communication on Online Platforms and the Digital Single Market Opportunities and Challenges for Europe'<https://ec.europa.eu/digital-single-market/en/news/communication-online-platforms-and-digital-single‑market‑opport
unities-and-challenges-europe> accessed on 10 June 2016.
 Commission, 'Digital Single Market – Commission updates EU audiovisual rules and presents targeted approach to online platforms' (Press release, 25 May 2016) <http://europa.eu/rapid/press-release_MEMO-16-1895_en.htm> accessed on 10 June 2016.
 Albert Breton, Competitive Governments: An Economic Theory of Politics and Public Finance (CUP 1998) 185.
 Wallace E Oates, 'Toward a Second-Generation Theory of Fiscal Federalism' (2005) 12 International Tax and Public Finance 349, 356-360.
 ibid, 351.
 Council (fn 3), Recital 34.
 Council (fn 3), Recitals 1 and 22.
 Loïc Azoulai, The Question of Competence in the European Union (OUP 2014) 102.
 Her Majesty's Government, 'Review of the Balance of Competences between the United Kingdom and the European Union. Competition and Consumer Policy Report' (2014) para 1.11.
 Council (fn 3), Recital 34.
 Patricia Vidal and Pablo Solano, 'Towards a More EU Consistent Approach To The Analysis of Vertical Restraints in the E-Commerce Sector' (2016) 2(2) Competition Law and Policy Debate 45, 46.
 CEAHR(fn 2), para 176.
 Lewis Crofts and Mari Eccles, 'Enforcers must understand the "rationale" behind online trade restrictions, Motta says' (MLex, 3 June 2016) <http://www.mlex.com/
GlobalAntitrust/DetailView.aspx?cid=800911&siteid=190&rdir=1> accessed on 7 June 2016.
 Francisco E González-Díaz and Matthew Bennett, 'The Law and Economics of Most-Favoured Nation Clauses' (2015) 1(3) Competition Law and Policy Debate 26, 27.
 Autorité de la Concurence 21 April 2015 15-D-06, para 56.
 Commission, 'Guidelines on Vertical Restraints' (Notice)  OJ C130/01, para 48.
 Ingrid Vandenborre and Michael J Frese, 'Most Favoured Nation Clauses Revisited' (2014) 35(12) ECLR 588, 592.
 Lear, 'Can "Fair" Prices Be Unfair? A Review of Price Relationship Agreements' (2012) OFT Report 1438, para 6.75 <http://www.learlab.com/pdf/oft1438_1347291420.
pdf> accessed on 8 May 2016.
 Justin P Johnson, 'The Agency Model and MFN Clauses' (2014) Cornell University ‒Samuel Curtis Johnson Graduate School of Management, 1 <http://ssrn.com
/abstract=2217849> accessed on 7 May 2016.
 Ariel Ezrachi, 'The Competitive Effects of Parity Clauses on Online Commerce' (2015) 55 Oxford Legal Studies Research Paper, 2 <http://ssrn.com/abstract=2672541> accessed on 7 May 2016.
 Commission (fn 21), paras 12-21.
 Michael L Weiner and Craig G Falls, 'Counseling on MFNs After E-books' (2014) 28(3) Antitrust Magazine 68, 71.
 Oxera, 'Most-favoured-nation clauses: falling out of favour?' (Agenda, November 2014) <http://www.oxera.com/Latest-Thinking/Agenda/2014/Most-favoured‑natio
n-clauses-falling-out-of-favour.aspx#_ftn6> accessed on 7 June 2016.
 Weiner and Falls (fn 27), 71.
 Jonathan B Baker and Judith A Chevalier, 'The Competitive Consequences of Most-Favored-Nation Provisions' (2013) 27(2) Antitrust Magazine 20, 24.
 Lear (fn 23), paras 6.49-6.50.
 Andre Boik and Kenneth S Corts, 'The Effects of Platform MFNs on Competition and Entry' (2016) JL & Econ (forthcoming) 19.
 Competition and Markets Authority, 'Report on the Private Motor Insurance Market Investigation' (2014) para 8.42.
 Lear (fn 23), paras 6.44-6.45 and 6.64-6.65.
 Bundeskartellamt 20 December 2013 B9-66/10, para 157.
 E-Books(Case COMP/39.847) Commission Decision 2013/C 378/14  OJ C378/25.
 Bundeskartellamt (fn 35).
 Commission (fn 21), para 102.
 Ezrachi (fn 25), 14.
 ibid, 15.
 Lear (fn 23), paras. 6.66-6.70.
 Competition and Markets Authority (fn 33), paras 8.62-8.63.
 According to Commission (fn 21), para 107(4), the hold-up problem is defined as the disincentive of the supplier or the buyer to undertake client-specific investments, such as in special equipment or training, before particular supply arrangements are fixed.
 Ezrachi (fn 25), 3-4.
 González-Díaz and Bennett (fn 19), 34.
 By horizontal free-riding we understand online platforms' benefiting from one another's investments by offering better conditions, whereas this phenomenon is of a vertical nature where the seller itself undercuts the prices quoted on the online platform.
 Martha Samuelson, Nikita Piankov and Brian Ellman 'Assessing the Effects of Most-Favored Nation Clauses' (American Bar Association Section of Antitrust Spring Meeting, March 2012), and Baker and Chevalier (fn 30).
 González-Díaz and Bennett (fn 19), 35-36.
 Konkurrensverket 15 April 2015 596/2013, Autorité de la Concurence (fn 20), and Autorità Garante della Concorrenza e del Mercato21 April 2015 I779.
 Bundeskartellamt, 'Bundeskartellamt issues statement of objections regarding Booking.com's "best price" clauses' (Press release, 2 April 2015) <http://www.bundes
kartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/2015/02_04_2015_Boking.html> accessed on 8 May 2016.
 Autorité de la Concurence (fn 20), para 122.
 ibid, para 123.
 ibid, paras 127-129.
 ibid, para 230.
 Konkurrensverket(fn 49), paras 21.
 ibid, para 25.
 ibid, paras 26, 27 and 30.
 Konkurrensverket 10 October 2015 595/2015, and Autorità Garante della Conco-rrenza e del Mercato 11 April 2016 I779.
 Bundeskartellamt (fn 50).
 Lewis Crofts, 'Booking.com's pricing clauses ruled illegal in Germany (update*)' (MLex, 23 December 2016) <http://www.mlex.com/GlobalAntitrust/DetailView.as
px?cid=800911&siteid=190&rdir=1> accessed on 7 June 2016.
 Matthew Newman, 'Booking.com loses bid to suspend German antitrust decision at Düsseldorf court' (MLex, 9 May 2016) <http://www.mlex.com/GlobalAntitrust/Detai
lView.aspx?cid=793030&siteid=190&rdir=1> accessed on 10 May 2016.
 Philippe Chappatte and Helen Townley, 'Online Hotel Bookings - A Joint European Approach or a Most Favoured Nation?' (Slaughter and May, Briefing, May 2015), 3 <https://www.slaughterandmay.com/media/2497093/online-hotel-bookings-a-joint-european-approach-or-a-most-favoured-nation.pdf> accessed on 8 May 2016.
 Crofts and Eccles (fn 18).
 Bundeskartellamt (fn 35), paras 188-195.
 Commission Regulation 330/2010 of 20 April 2010 on the application of Article 101(3) TFEU to categories of vertical agreements and concerted practices  OJ L102/1.
 Bundeskartellamt (fn 35), para 187.
 Vandenborre and Frese (fn 22), 591.
 Autorità Garante della Concorrenza e del Mercato (fn 58), para 57.
 Konkurrensverket(fn 49), para 20, Autorità Garante della Concorrenza e del Mercato (fn 49) para 7, Autorité de la Concurence(fn 20),para 7, and Bundeskartellamt (fn 35), para 10.
 For all, Konkurrensverket(fn 49), para 25, and Bundeskartellamt(fn 35), para 87.
 Competition and Markets Authority 'CMA closes hotel online booking investigation' (Press release, 16 September 2015) <https://www.gov.uk/
government/news/cma-closes-hotel-online-booking-investigation> accessed on 8 May 2016.
 Competition and Markets Authority (fn 33), para 8.123.
 Case T-286/09 Intel v Commission ECLI:EU:T:2014:547.
 Case C-23/14 Post Danmark A/S v Konkurrencerådet ECLI:EU:C:2015:651.
 Groupement des Cartes Bancaires(fn 1), para 58, and Maxima Latvija (fn 1), para 17.
 Case C-74/14, ECLI:EU:C:2016:42, 'Eturas' UAB v Lietuvos Respublikos konkure-ncijos taryba.
 Commission Notice on agreements of minor importance which do not appreciably restrict competition under Article 81(1) of the Treaty establishing the European Community (de minimis)  OJ C368/13, para 11(2).
 Case C-226/11, ECLI:EU:C:2012:795, Expedia.
 Commission (fn 21), para 48.
 Konkurrensverket (fn 49), para 18.
 Steven C Salop and Fiona S Morton, 'Developing an Administrable MFN Enforcement Policy' (2013) 27(2) Antitrust Magazine 15, 17.
 Richard Whish and Brenda Sufrin, 'Article 85 and the Rule of Reason' (1987) 7(1) YBEL 1.
 Commission, 'Guidelines on the application of Article 81(3) of the Treaty' (Notice) SEC (2004) OJ C101/97, para 11.
 Expedia(fn 78).
 ibid, para 16.
 Commission (fn 21), para 48.