CAT among the pigeons? How does Compare The Market’s successful appeal impact the approach to MFN clauses under competition law?

What are MFN or price parity clauses?

MFN or price parity clauses typically involve a provider agreeing to offer rights and benefits to a contracting party that are equal to or more favourable than rights and benefits offered to other parties or on other sales channels. This obligation may be “narrow” (limited to the provider’s direct sales channels) or “wide” (market-wide, or at least competing indirect sales channels). Using home insurance and price comparison websites as an example (given the nature of the Compare The Market case):

(a) a narrow MFN would involve an insurance provider agreeing not to offer lower prices on its own website (or other direct sales channels) than the prices quoted for its products on a price comparison website, but leave the insurance provider free to offer lower prices on other price comparison websites or other indirect channels; whereas

(b) a wide MFN prevents the insurance provider undercutting prices quoted for its products on a price comparison website not only on its own website but also on any other price comparison websites or other sales channels.

As explained below, the distinction between narrow and wide MFNs is an important one for competition law purposes.

How do the EU and UK Vertical Block Exemptions treat these types of clauses?

The EU and UK Vertical Block Exemptions provide a “safe harbour” for vertical agreements (i.e. agreements between parties at different levels of supply) that meet certain requirements. Such agreements are deemed compatible with competition law, whereas agreements that do not meet the requirements must be assessed to determine whether they are compatible or problematic.

Under the approach which applied until 31 May 2022, all MFN/parity clauses could be block-exempted. However, decisional practice of various competition authorities in recent years, including the CMA, suggested that wide MFNs/parity clauses may be anti-competitive. The new EU VBER and UK VABEO applying from 1 June 2022 introduce much-needed clarity and revise the approach in relation to such clauses. Importantly, they have also introduced divergence between the approaches in the EU and the UK, whereas the EU and UK approaches were previously aligned.

In the EU, under the new VBER, wide retail parity clauses imposed by online platforms (referred to as across-platform parity clauses) causing a buyer of online intermediation services not to offer, sell or resell goods or services to end-users under more favourable conditions via competing online intermediation services are now “excluded” restrictions, meaning they are not block-exempted and their compatibility with Article 101 of the Treaty on the Functioning of the European Union (TFEU) must be individually assessed. However, provided any across-platform parity clause can be severed from the rest of the agreement, the remainder of the agreement can continue to be block-exempted. 

The UK has adopted a stricter approach than the EU. Under the new VABEO, all wide retail parity clauses (not only those imposed by online platforms) are considered “hardcore” restrictions. Hardcore restrictions are generally viewed as serious restrictions of competition, and (in contrast to the approach for excluded restrictions) the inclusion of a hardcore restriction in an agreement means the entire agreement loses the benefit of the block exemption.

Both the new EU and UK approaches focus on wide parity clauses in the retail context. Wide parity clauses at other levels of the supply chain continue to benefit from the block exemptions as they are considered less likely to affect competition.

Narrow parity clauses also continue to be eligible for block exemption. However, the respective guidelines accompanying the new EU VBER and UK VABEO highlight that in certain circumstances narrow retail parity clauses may also have restrictive effects, effectively indirectly producing restrictive effects equivalent to those produced by wide retail parity clauses.

What happened in the Compare The Market case?

On the back of the CMA’s market study into digital comparison tools, the CMA opened an investigation into Compare The Market in 2017. Compare The Market had contracts with 32 insurance providers which contained wide MFNs – preventing the insurance providers from offering lower prices on other comparison websites and protecting Compare the Market from being undercut. 

After a three-year investigation, the CMA fined Compare The Market £17.9 million in November 2020, having concluded that its wide MFNs infringed both the Chapter I Prohibition under the UK Competition Act 1998 and Article 101 TFEU, which prohibit agreements that are anti-competitive “by object” (i.e. by their very nature) or “by effect”. Significantly, the CMA pursued an effects case, meaning they needed to demonstrate that Compare The Market’s wide MFNs had an appreciable anti-competitive effect (whereas there is no need to prove anti-competitive effects if an agreement is anti-competitive by object).

The CMA defined the relevant market as the provision of price comparison website services for home insurance in the UK, and specified this was a two-sided market (given price comparison websites match two user groups, in this case consumers and home insurance providers), comprising the supply of: (a) customer introductions services to insurers; and (b) price comparison services to customers. The CMA found that the network of wide MFNs had an appreciable effect on competition between price comparison websites and between insurance providers competing on such websites by:

(a) reducing price competition between price comparison websites;

(b) restricting the ability of Compare The Market’s rivals to expand, enabling Compare The Market to maintain or strengthen its market power; and

(c) reducing price competition between insurance providers competing on price comparison websites.

Why was Compare The Market successful on appeal?

Compare The Market sought to challenge the CMA’s decision on a number of grounds before the CAT, and was ultimately successful in persuading the CAT to overturn the CMA’s decision. Crucially, in its judgment published on 8 August 2022, the CAT concluded that the CMA had: (a) adopted a flawed market definition; and (b) failed to provide evidence of anti-competitive effects on premiums, commissions or promotional deals, failed to establish the counterfactual and causation, and also made further factual errors regarding Compare The Market’s wide MFNs.

Market definition

Market definition is a key tool in assessing the importance of the existence of anti-competitive effects, and was complex in this case due to the two-sided nature of the market, with novel elements and practice developing regarding the approach to two-side markets. However, the CAT concluded that the CMA’s market definition was “unreal”, “materially wrong” and “not fit for purpose”.

The CAT considered that the CMA had fallen into the trap of adopting a market definition that leads to a particular outcome – whereas the approach to market definition ought to be outcome neutral and not pre-determine a finding of anti-competitive effect.  In terms of the CMA’s specific approach to defining the market, the CAT also found that the CMA adopted an inaccurate definition of the consumer side of the market, failed to properly consider the significance of other channels for the purchase of home insurance by consumers, failed to properly test for demand substitutability on the consumer side of the market, and incorrectly included narrow MFN clauses that were prevalent in the market in the market definition assessment.

The CAT therefore concluded that the market definition needed to be redone (which the CAT did) and this weakened the CMA’s decision. However, this was not necessarily fatal to the CMA’s finding of an infringement.

Anti-competitive effects, counterfactual and causation and factual errors

What was fatal to the CMA’s case was the underlying evidence that the CMA had relied upon. The CAT concluded there was “no reliable evidence” that Compare The Market’s wide MFNs had any adverse effect on premiums or commissions, and considered it unlikely that the wide MFNs had any effect on maintaining premiums or commissions at a higher level than they would otherwise have been.

The CAT considered that a great deal of the CMA’s analysis operated at the level of theory or bare assertion, with no significant reference to quantitative evidence. The CMA’s decision essentially relied on qualitative evidence (documentary evidence), but the CAT considered this evidence was “anecdotal” (of very little weight) and lacked depth and consistency with the CMA’s theory of harm (including two instances where the wide MFNs caused premiums to fall, not rise). More seriously, this evidence was untestable by both Compare The Market and the CAT, with it unclear what evidence the CMA was actually relying on. Overall, this material was insufficient to meet the burden on the CMA to establish an infringement.   

The CAT also found a number of other factors undermined the CMA’s case, the first being that the CAT does not consider wide MFNs are necessarily effective save in the most egregious of cases. Apparent infringements of the wide MFNs evidenced the limits of their effectiveness (even though these were contractually binding and rigorously enforced), suggesting that wide MFNs would only be effective in limited cases where an insurance provider went out of their way to discriminate against Compare The Market by “loading” the premiums quoted on that price comparison website and no other.  

The CAT also found it concerning the CMA did not take account of the effect of prevalent narrow MFNs in the market, i.e. whether obligations on insurance providers not to quote premiums on their direct sales channels that undercut their premiums on price comparison websites might produce a similar effect to the wide MFNs.  If there were such an effect this would undermine the CMA’s theory of harm, but the CMA did not address this. 

Other relevant factors included that the CAT did not consider the theoretical argument against wide MFNs to be particularly strong in the markets under consideration; the CMA’s decision materially understated downward pressure on premiums and commissions on the consumer side of the market (by getting the market definition wrong in this respect); the CMA failed to consider the significant competition between home insurance products that existed through different routes  (in particular inter-brand competition, i.e. between products of different insurance providers); and quantitative evidence provided by Compare The Market was broadly consistent with the CAT’s views on the theoretical operation of the market and the value of the qualitative evidence.   

In relation to promotional discounts, the CAT did not consider that the competitive structure of the market was harmed, even potentially, through an effect on promotional discounts.

How does Compare The Market’s appeal affect the approach under the new VABEO?

The CAT’s findings arguably appear at odds with the new approach to wide retail MFN/parity clauses under the VABEO whereby such clauses are now categorised as hardcore restrictions, i.e. serious restrictions (noting that most hardcore restrictions are also “by object” restrictions). However, the CMA did not pursue Compare The Market’s wide MFNs as by object restrictions, and while the CAT agreed with the CMA on this point this remains an issue that has not been thoroughly tested. Nevertheless, in addition to finding that the evidence relied on by the CMA was inadequate to demonstrate anti-competitive effects, the CAT doubted the effectiveness of wide MFNs as well as the theoretical argument against such restrictions in the markets under consideration, which appears to cast further doubt on their classification as hardcore restrictions.   

It is unfortunate that the CAT’s judgment does not mention the new approach to wide retail MFNs under the VABEO, and it remains to be seen whether the CMA will seek to appeal the CAT’s judgment (the CMA has until mid-September to do so). Possibly any such further appeal may help clarify the position but it appears that this will remain a vexed area of law for the time being.¹ In any event, there is no suggestion that the VABEO will be revised in the near term and it therefore remains the case that parties who wish to ensure their vertical agreements are block exempted under the VABEO should refrain from including wide retail MFNs in their agreements.