A couple of questions struck me as I read the puzzling clearance decisions adopted by the Swedish Competition Authority in Coop/Netto and NEP/HDR. One question is: have we entered an era of lenient Swedish merger control? Another is: isn’t it time to introduce a possibility for third parties to appeal Swedish clearance decisions? In this post I discuss the food retail merger Coop/Netto and the TV production merger NEP/HDR and then raise 4 questions we should be discussing on the state of Swedish merger control.*
In 2019, the second largest Swedish food retail chain, Coop, acquired all of Netto's Swedish food stores
Coop's acquisition of Netto's hard discount stores
The food discounters enters Sweden
In the early 2000’s, two European hard discount food retailers, Netto (Danish) and Lidl (German), entered the Swedish market. The Swedish food retail market was then one of the most concentrated in Europe and some may have believed that Netto and Lidl had the potential to start stirring the pot of Swedish food retail by challenging the hegemony of Ica, Coop and Axfood.**
At face value, this looked like a transaction destined for phase II
Almost 20 years later it was clear that while both Netto and Lidl had spiced things up by rolling out a significant number of stores (around 170 each), they had not cracked the nut of concentration in the Swedish food retail market. In 2018, the SCA concluded in a report that the Swedish food retail market remained one of the most concentrated in all of Europe (see the SCA report 2018:4), and that the three giants Ica, Coop and Axfood still controlled nearly 90% of the food retail market on a national level.
Against this backdrop, it was announced in early summer 2019 that Coop, the second largest food retail chain was acquiring all of Netto’s 163 Swedish stores. In other words: the number two in a highly concentrated consumer market was buying a (comparatively) recent entrant with an aggressive pricing strategy. At face value, this looked like a transaction destined for phase II. However, the transaction sailed through unconditionally in phase I (dnr. 344/2019, press release and decision available in Swedish here). Let’s look at how.
Like this post? Join my mailing list to support this type of content.
The Swedish Competition Authority ("SCA") identified concerns but considered them small potatoes
Predictably, the SCA relied on a market definition in line with fairly established decisional practice from the European Commission as well as various national competition authorities. Under this definition of the relevant market, all categories of food retailers (except for service stations and convenience stores, such as 7-Eleven, Pressbyrån or gas stations) are considered to compete with each other, as long as they have stores located within the same 5-10 minute drive time catchment area. In other words, one of Netto’s hard discount stores is considered a competitor to one of Coop’s supermarkets, provided that they are within 5-10 minutes driving time from each other.
This means that the SCA had to analyse the competitive conditions in a high number of local markets. In this analysis, the SCA found that Coop’s acquisition of Netto’s Swedish stores could lead to competition concerns in a total of 20 local markets. The SCA also found that there was a high risk of competition concerns in 2 of these local markets. What this means more explicitly is that the SCA concluded that consumers in a total of 20 local markets, due to this transaction, could face for instance higher food prices, reduced availability or quality.
"a significant risk of anticompetitive effects"
The SCA on two of the local markets
Having established these concerns, the SCA then dismissed them in a remarkably brief conclusion. The SCA noted that the local markets where significant concerns had been identified constituted only a minor segment of the total food retail revenues in Sweden. The SCA also noted that some local markets where competition concerns had been identified did exhibit characteristics that could mitigate concerns. Therefore, the SCA did not consider that the requirements for the SCA to legally intervene were met and cleared the transaction.
Retail mergers are normally assessed based on local market competition
Food for thought
From an outside perspective, I do find it a bit curious that the SCA considered it appropriate to clear this transaction already in phase I, despite the fact that the SCA had identified competition concerns in a fairly high number of local markets. It is also striking how summarily the SCA dismissed the concerns and how little insight the decision gives into the SCA’s substantive reasoning. Only a few paragraphs of the decision are spent on the relevant effects and excluding boilerplate decision text and factual information, the decision is actually only a few pages long. The brevity of the decision relating to competitive effects makes it almost impossible for outside observers to examine the arguments or to find any sort of legal guidance.
I understand that this may be dismissed as quibble on form, but I think it is important to consider this criticism in light of the enormous stakes at play when assessing competition in the food retail market. These stakes include the following:
Few markets are as socioeconomically important as the food retail market
Food purchases represent the second largest expenditure item for Swedish households (the largest is cost of living, such as electricity, heat etc., according to data compiled by Ekonomifakta). Food is also something every household needs to buy and the lower the household income, the higher the share of spending on food. Increases in food prices affect every household, and disproportionally affect the purchasing power of lower income households (which is why Sweden and many other countries have adopted lower VAT rates for food).
Ideally, the socioeconomic importance of the food retail market should instill a heightened sense of mission at the SCA, prompting a particularly diligent scrutiny when reviewing food retail cases and also a desire to explain the details of what goes into the decisions. None of this is displayed in the SCA’s Coop/Netto decision.
Swedish food prices are already among the highest in the world
In 2018, the SCA reported that Swedish food prices are among the highest in the world (SCA report 2018:4). Granted, as a percentage of average income, the food prices in Sweden are not as comparatively high. The problem though is that food is an essential good and that higher food prices disproportionally affect low income households. This means that the comparatively high food prices could be a significant problem, especially in light of rising income inequality in Sweden. If the comparatively high prices are a problem caused by insufficient competition in the Swedish food retail sector, it is a problem within the purview of the SCA.
The Swedish food retail market was already highly concentrated
As mentioned, the SCA concluded in the 2018 report (SCA report 2018:4) that the Swedish food retail market is one of the most concentrated in all of Europe (CR3 nearly 90% and HHI of nearly 3300). Through this transaction, it became even more concentrated. Of course, market concentration is not in itself necessarily negative for competition outcomes, and as explained, food retail markets are normally assessed locally, not nationally. Still, given that Coop/Netto did create horizontal overlaps also locally and that the SCA identified potential concerns due to these overlaps, it seems a bit curious that the SCA did not discuss how these competition concerns on a local level could relate to the already high degree of concentration on a national level.
The hard discount stores may be particularly important for local prices
The SCA noted early in the decision that hard discount stores such as Lidl and Netto exert a competitive price pressure on all other types of food retailers, also on a local level. In its 2018 report, the SCA also discussed how both Swedish and international research showed that food prices increase as distances to competing stores increase. This is not completely reconciled in the decision’s competitive analysis.
Some may find it problematic that the SCA, in light of these stakes, cleared Coop/Netto already in phase I despite having identified competition concerns and without providing more than the bare minimum of a motivation in the decision. Others may think the SCA has bigger fish to fry on its limited budget and should apply a high materiality threshold.
Let’s compare Coop/Netto with another clearance decision from this year, NEP/HDR, a case without the same socioeconomic weight as Coop/Netto but that also demonstrates a lenient approach to identified competition concerns.
NEP's acquisition of rival TV producer HDR
Setting the scene
In order for viewers to be able to watch live video from events such as sport matches, there must be someone actually producing the content. I.e. someone needs to be where the event takes place to set up cameras, draw cables, point microphones and record to produce the actual signal. NEP and HDR are two companies in Sweden that do exactly this; provide outsourced production services for different types of live events.
During the summer of 2019, NEP announced that it was acquiring HDR. The transaction was filed with the SCA that opened a phase II review and then cleared the transaction unconditionally almost 2 weeks ahead of the phase II deadline (dnr. 435/2019, press release and decision available in Swedish here).
The NEP/HDR transaction concerned the markets for outsourced TV production
The SCA identified massive overlaps, but no showstoppers
Without taking a definitive position on the relevant market, the SCA concluded that this was a transaction between two market leading close competitors. No matter which market definition the SCA would rely on, NEP and HDR would be number 1 and number 2, with a combined market share that indicated the creation or strengthening of a dominant position, with small and weak competitors. The SCA could also conclude that NEP and HDR are each other’s closest competitors, competing head on in the absolute majority of projects that are exposed to competition. From a merger control perspective it sort of doesn’t get much worse than this. Nevertheless, the SCA did find a way to approve this transaction.
"The combined market shares on all potential markets are on a level indicating the creation or strengthening of a dominant position"
The SCA on the merging parties
According to the SCA, potential competition from foreign production companies would prevent the merger between NEP and HDR from having anticompetitive effects on the Swedish market. The SCA’s support for this conclusion included for instance that two foreign companies were assessing the possibility of taking on Swedish productions in the future. The SCA also referred to foreign production companies previously taking on projects such as producing the World Cup in orienteering and the Swedish open. Furthermore, the SCA concluded that Swedish customers previously had turned to foreign production suppliers to an extremely limited extent, but assumed that this would change if the SCA would approve the number 1 and the number 2 on the market to merge. According to the SCA, the mere threat of entry by foreign companies would discipline NEP and HDR.
A few critical angles
Reading the decision it is easy to get the impression that the SCA, for some reason, decided early on to clear this transaction and then went out of its way to find a reason to do so. Potential competition and market entry are of course important scenarios that must be considered. But for market entry to determine the outcome of a merger review, the loss of competition due to the transaction must be assessed and then compared with not only the likelihood that market entry actually will occur but also the likelihood that such entry sufficiently and timely could mitigate the potential consequences of the identified loss of competition (see section IV of the European Commission’s horizontal merger guidelines). Some may question whether the SCA adequately covered the relevant steps of this analysis in the NEP/HDR decision.
The SCA is banking on market entry to prevent competition concerns due to NEP/HDR
The SCA did indirectly conclude that the transaction would lead to a massive loss of competition, since the parties were market leaders and close competitors, but did not discuss or clarify what the anticompetitive consequences of this loss could look like absent mitigating factors. The SCA did also discuss the likelihood of market entry, but did not discuss the likelihood that such entry actually could mitigate any concerns due to the loss of competition. For example, the SCA does not seem to have taken into account the competitive conditions affecting the foreign companies or the fact that the acquiring party, NEP, claims to be a worldwide leader within outsourced production.
In my view, many of the factors that the SCA pointed to as indicating a high likelihood of future entry seemed vague, bordering far-fetched.
In my view, many of the factors that the SCA pointed to as indicating a high likelihood of future entry seemed vague, bordering far-fetched. Referring to the World Cup in orienteering, hardly an important event for broadcasters or production companies, as an example of foreign production companies previous activities in Sweden, illustrates how the SCA must have been hard-pressed to find any such examples at all.
As with the Coop/Netto decision, some may find the NEP/HDR decision outrageously lenient and others may find it reasonable. In contrast with the Coop/Netto transaction, this transaction was at least referred to phase II and it could be argued that the NEP/HDR transaction concerned markets of much less socioeconomic importance than Coop/Netto. However, even if you would consider it appropriate for the SCA to be more lenient towards markets of arguably lower socioeconomic importance, you probably also agree that transparency is important and that the SCA should not be taking policy or prioritisation considerations into account surreptitiously.
4 critical questions on the state of Swedish merger control
Do the clearance decisions in Coop/Netto and NEP/HDR indicate a more lenient approach from the SCA?
As perhaps is evident from the above, I personally consider the SCA’s assessments in these two cases puzzlingly lenient. However, it is probably not possible to draw any conclusions at all as to whether they indicate an overarching trend in Swedish merger control. The SCA did during 2019 block a merger in the dairy industry (dnr. 661/2018, press release and decision, appeal pending) and it is difficult to discuss any trends on basis of only two decisions. This being said, it will be interesting to observe the SCA’s approach, priorities and decision drafting during 2020.
Would the European Commission or any other European competition authority ever be as lenient as the SCA was in these cases?
I believe few other European competition authorities, and especially not the European Commission, would have sided with the SCA, if they were to review two transactions with corresponding facts. At least it is likely that other authorities would put up more of a fight and not be as dismissive of identified concerns.
Conventional wisdom has for a long time suggested that it often is wise to steer transactions towards the SCA (when possible) rather than European Commission, partly because of the vastly more efficient procedural approach of the SCA. If the SCA’s lenient approach in these two transactions would be indicative of the SCA’s substantive approach going forward, the conventional wisdom of preferring the SCA before the European Commission would be truer today than ever.
Should the SCA adopt more thorough and detailed clearance decisions, even if this would risk dampening the SCA's overall efficiency?
Even if you would disagree with me that the SCA displayed a lenient substantive approach in these cases, and even if you disagree with me that this approach stands out from a European perspective, you may perhaps agree with me on this: the SCA should seek to better explain the considerations that go into its decision making.
As mentioned, the SCA is generally considered to be a procedurally efficient and fast competition authority. The SCA managed to clear Coop/Netto unconditionally already in phase I despite significant horizontal overlaps and identified competition concerns in a high number of local markets. The SCA managed to clear NEP/HDR unconditionally almost 2 weeks ahead of the phase II deadline expiry. The SCA usually clears non-controversial mergers within only a few working days form filing.
While impressive, the SCA’s efficiency may come with drawbacks. Adopting decisions this quickly may require the SCA to be comparatively laconic in its written decisions. Drafting good decisions takes time. Substantively assessing mergers also takes time. But sometimes legal decision making must be allowed to take time in order to provide citizens and the market with legal guidance and improve legal certainty but also to maintain transparency in the exercise of public power.
Should Sweden introduce an ability for directly concerned third parties to challenge SCA clearance decisions in court?
The Swedish Competition Act allows merging parties to appeal blocking decisions by the SCA. However, the act does not allow directly concerned third parties to appeal clearance decisions. In some jurisdictions it is possible for third parties to appeal the competition authority’s clearance decisions. For example, it is possible for directly concerned third parties to appeal clearance decisions adopted by the European Commission.
Introducing the ability for third parties to appeal the SCA’s clearance decisions would significantly alter the balance of power during during Swedish merger reviews. Instead of having to worry only about whether a blocking decision would stand on appeal by the merging parties, the SCA would have to worry also about whether a clearance decision would stand on appeal by a third party. In other words, any “threats” from merging parties regarding their intention of appealing a potential blocking decision would be balanced out by countervailing threats by third parties regarding their intention of appealing a potential clearance decision. I personally believe this could put some healthy boundaries on how far the SCA can go to clear transactions.
* My intention with this post is not to criticise the Swedish Competition Authority for making incorrect substantive assessments. As an outside observer it is impossible to have the same level of insight into the relevant markets as the SCA’s case officers, who presumably reviewed all relevant facts and evidence in close detail. My intention with this post is merely to apply some healthy skepticism and to raise a few questions relating to the SCA’s decision-making in merger review cases.
Where appropriate, I have translated relevant quotes from the decisions from Swedish to English. Any opinions in this post cannot be attributed to anyone but me.
In this post I use the term “clearance”. However, strictly speaking, the SCA does not “clear” transactions, it only holds in a decision that it will not take any action in relation to the transaction.
** When Lidl first entered Sweden it did so through a joint venture partly owned by ICA.