Resale Price Maintenance and Artificial Intelligence

September 17, 2018

With artificial intelligence, and in particular pricing algorithms, under increasing scrutiny from worldwide
competition regulators, the European Commission’s latest sanctions against
four electronic consumer manufacturers for engaging in online resale price
maintenance – the first such decisions adopted by the Commission, and its first
‘resale price maintenance’ decisions in 15 years – shed important light on this
topic. In particular, it confronts the issue of vertical restraints implemented
by some manufacturers in the EU, at times in only a few national markets, whose
effects are compounded by price comparison tools and pricing algorithms.

On 24 July 2018, the Commission
sanctioned four electronic consumer manufacturers
– Asus, Denons &
Marantz, Philips and Pioneer – for engaging in ‘fixed or minimum resale price
maintenance’ by restricting the ability of online retailers to set their own
retail prices for widely used consumer electronics products, including tablets,
headphones, speakers and kitchen appliances. Commissioner Margrethe Vestager
declared: ‘As a result of the actions taken by these four companies, millions
of European consumers faced higher prices for kitchen appliances, hair dryers,
notebook computers, headphones and many other products. This is illegal under
EU antitrust rules’.

The millions of consumers mentioned by Commissioner Vestager
are, however, not only online shoppers of the Asus, Denons & Marantz,
Philips and Pioneer products referred to in the Commission’s decisions; they
could also include purchasers of similar products of different brands. Indeed,
as a result of the vast majority of online retailers using pricing algorithms,[i] which automatically
adapt retail prices to take account of those of competitors, the pricing
restrictions unlawfully imposed by the four infringing companies on low-price
online retailers would have had an impact on overall online prices for similar
consumer electronics products of other manufacturers.

The potential facilitating role of algorithms on
anticompetitive conduct, used for instance to automatically fix prices,
horizontally (ie cartel behaviour) or vertically (ie RPM) has, with good
reason, become one of the hottest topics in EU competition law over the past
few years, in particular regarding the question of the responsibility of the
company controlling the algorithm.[ii] But the Commission’s recent decisions now
seem to point in another direction: the effects of RPM is aggravated by competitors’
pricing practices, and has become a focus of EU competition authorities.

Thinking a step further, if full redress is to be provided, consumers’
right to seek redress and obtain damages for the harm caused by such
anticompetitive behaviour should capture the widespread effect on the market.

Vertical restraints under EU competition rules

Under EU competition rules, manufacturers should not take
any actions that interfere with the freedom of retailers to set their final
prices to customers by establishing a fixed or minimum resale price or price
level required to be observed by the retailers. Such agreements are deemed to
be a restriction of competition ‘by object’ (akin to a ‘per se’ infringement)
under Article 101(1) of the Treaty on the Functioning of the European Union,
and a hard-core restriction within the meaning of Article 4(a) of the
Commission’s Vertical Block Exemption Regulation. They are subject to a strict
assessment of possible exemptions as regards market efficiency, market share
thresholds, and other situations where a pricing recommendation does not amount
to a minimum or fixed resale price because of threats, pressure or incentives.[iii]

A number of competition authorities across Europe have fined
manufacturers for RPM practices, including the German Bundeskartellamt, which
imposed fines in 2015 and 2016 on 27 companies in the food retail sector for
illegally fixing retail prices,[iv] and the UK’s Competition and Markets
Authority, which issued a decision in May 2017 finding that several businesses
in the light-fittings sector infringed competition law as a result of the
setting of minimum prices for online sales.[v] But, until now, the Commission has
typically left such enforcement action to the national competition authorities.
The sanctions against Asus, Denons & Marantz, Philips and Pioneer were its
first RPM decisions in 15 years; the last fine imposed by the Commission to
sanction RPM dates back to 2003, in a completely different economic environment
with a focus was on bricks-and-mortar (rather than online) retailers.

In the recent Commission decisions, the manufacturers used
sophisticated monitoring tools, namely pricing algorithms, to effectively track
resale price setting in the distribution network, and to intervene swiftly in
case of pricing changes. Online retailers offering the products at the lowest
prices were specifically targeted by the practices, and faced threats or
sanctions such as having their supplies blocked if they sold the applicable
prices below the prices required by the manufacturers. The scope of practices
sanctioned differed across all four manufacturers. While the bulk of the
practices took place between 2011 and 2013, in some instances the conduct
continued until 2015, and spanned a number of countries.

The role of pricing algorithms in vertical restraints

Pricing software, being a powerful tool for manufacturers
and retailers to frequently, if not instantaneously, monitor online prices,
also makes it easier for manufacturers to detect deviations from their pricing
requirements and retaliate accordingly, thus contributing to the effectiveness
of RPM practices and ensuring a more stable and durable effect on prices.

Interestingly, the algorithms not only facilitated the RPM
practices. Similar software used by competitors also contributed to intensifing
the effects of such practices, affecting the market prices more generally. This
element was emphasised by Commissioner Vestager:

‘In fact, by targeting specific low price retailers, the
four manufacturers were also able to influence the prices other online
retailers charged. This is because online retailers use pricing algorithms and
price comparison websites to constantly monitor, in real time, the prices
charged by their competitors. They then adjust their prices accordingly. So if
one retailer offers lower prices than others, this prompts other competitors to
lower their prices. Conversely, if that retailer puts its prices back up, others will follow.

EU competition enforcers and the courts have already
positioned themselves regarding the role of algorithms as regards horizontal
agreements, which may facilitate or strengthen collusion among retailers,
allowing and accelerating the detection of deviations.[vi] But the role of algorithms is two-fold:
not only does it facilitate unlawful RPM, but it also contributes to spreading
the effects of a vertical restraint of competition.

In its July 2018 decisions, the Commission did not, at least
in the press release published to date, touch upon any assessment or
quantification of the effects on the market or consumer welfare, although
Commission Vestager did note the immediate effect on consumers that the RPM
practices would have had. This raises questions as regards private enforcement,
where consumers have suffered harm on a large scale and may want to seek
redress accordingly. Specifically, to what extent is the unlawful conduct
confined to the prices charged by the infringing undertakings for the products
in question; what of the pricing algorithms that could have compounded the
effect of the RPM across the market; how can consumers be redressed for the
higher prices they will have paid as a result?

A parallel with the ‘umbrella pricing’ theory?

In the context of horizontal agreements, the European Court
of Justice has endorsed the validity of a causal relationship between a cartel
and ‘umbrella’ pricing, namely inflated prices charged by non-cartelists whose
prices are benchmarked against market-wide prices which are artificially
inflated as a result of a cartel. The ECJ held in this context that: ‘even if
the determination of an offer price is a purely autonomous decision, taken by
the undertaking not party to a cartel, it must none the less be stated that
such decision has been able to be taken by reference to a market price
distorted by that cartel and, as a result, contrary to the
competition rules’.

This begs the question as to whether, if consumers are able
to claim compensation from cartelists for inflated prices paid to both
cartelists and non-cartelists, by analogy they may also be able to claim
compensation from the infringing companies for competitors’ similarly inflated
prices arising from the RPM practices whose effects are compounded by
algorithms. Such a case has yet to be brought.

‘E-commerce brings its own challenges for competition’

These issues are all the more important in the context of
practices targeting specific products in specific countries. The Commission has
also expressed concerns on market segmentation, which can be the consequence of
technical measures such as geo-blocking (also addressed in the description of
the Pioneer unlawful practices), and has opened several investigations
following on from the e-commerce sector inquiry Final Report adopted on 10 May

In that context, while e-commerce has lowered national
borders and contributed to the growth of cross-border trade within the EU (and
beyond), technological progress may also facilitate implementation and
monitoring of vertical (and horizontal) restrictions that may be contrary to
the principles of EU law.

A joint approach to tackle fast-evolving challenges

The nature of pricing algorithms and, more importantly, of
self-learning algorithms, coupled with endless scientific progress, poses
challenges to EU competition authorities not only to detect but also assess the
potential harm arising from practices involving artificial intelligence.
Accordingly, the fast-evolving nature of the digital markets requires a
pro-active, flexible, and creative approach to competition law enforcement, at
all levels.

This involves, for example, close cooperation between
competition enforcers in the development of their understanding of the
challenges raised by algorithms, as is demonstrated by the joint project
launched in June 2018 by the French Autorité de la concurrence and the German
Bundeskartellamt on algorithms and their implications on competition, seeking
to identify conceptual approaches to meet the challenges raised by algorithms;[viii] and the CMA’s recent
appointment of a new data unit
‘to better understand the impact that data,
machine learning and other algorithms have on markets and people’.

Apart from joint forces amongst competition enforcers,
noteworthy initiatives have been developed beyond the sole prism of competition
law, for instance with the development of a larger legislative toolkit to
create a European Digital Single Market.[ix]  This involves,
among many other things, the concepts of data privacy and other consumer
rights. It also involves the relationship between competition enforcers and
private businesses, such as the CMA’s Screening
for Cartels tool
, which provides the government enforcer with the tools to
detect and defeat cartels, using algorithms to spot unusual bidder behaviour
and pricing patterns which may indicate that bid-rigging has taken place, and
therefore potentially defeat algorithms with competing algorithms.

Anna Morfey is a partner in the London office of Hausfeld

Amandine Gueret is an associate in Hausfeld’s London and
Paris offices.

This article first appeared in Hausfeld’s quarterly client

[i] European Commission, Final
report on the E-commerce Sector Inquiry
, COM(2017) 229 final, Brussels, 10 May 2017: ‘A majority of
retailers track the online prices of competitors. Two thirds of them use
automatic software programmes that adjust their own prices based on the
observed prices of competitors

[ii] OECD, Directorate For Financial And Enterprise Affairs
Competition Committee, Algorithms
and Collusion – Note from the European Union,
14 June 2017,
DAF/COMP/WD(2017)12, p 9.

[iii] Commission Regulation (EU) No 330/2010 of 20 April 2010 on
the application of Article 101(3) of the Treaty on the Functioning of the
European Union to categories of vertical agreements and concerted practices.

[iv] Bundeskartellamt, Vertical
price fixing – Düsseldorf Higher Regional Court raises fine against drugstore
chain Rossmann, 
1 March 2018.

[v] CMA, Online
resale price maintenance in the light fittings sector
, Case 50343, 3 May

[vi] In that vein, the CMA fined online sellers of poster and
frames for automated collusion, through automated re-pricing software to
monitor and adjust the parties’ prices, on Amazon’s UK website (
Online sales of posters and
, Case 50223, 12 August 2016), and the CJEU confirmed that
companies cannot escape liability where collusion has been achieved and
executed through automated systems (Case C-74/14, Judgment of 21 January 2016 
— ‘Eturas’ UAB and Others v
Lietuvos Respublikos konkurencijos taryba).

[vii] Case C-557/12, Judgment of the Court of 5 June 2014, Kone AG and others v
ÖBB-Infrastruktur AG.

Autorité de la concurrence, ‘The French Autorité de la, 19
June 2018.

[ix] The
creation on 14 June 2018 of a new EU High Level Group on Artificial
Intelligence involving 52 experts from across industry, business and civil society
aiming at making recommendations on how to address mid-and long-term challenges
and opportunities related to artificial intelligence, including from a legal
perspective, is a further step towards a proactive approach to digital