Parties to an M&A transaction or parties that work together sometimes agree not to compete with each other (temporarily or otherwise). Such a non-compete benefits the party for which it is agreed, because it protects against competitive activities of the other party.
Under certain circumstances, non-competes may restrict competition because parties agree among themselves in which markets they will or will not be active, thereby effectively allocating markets. Depending on the market circumstances, market allocation between competitors is prohibited by the cartel prohibition set out in Article 101 of the Treaty on the Functioning of the European Union (TFEU) and in Article 6 of the Mededingingswet (Competition Act).
It is apparent from established case law (Remia and Pronuptia) that anti-competitive agreements are in any event permitted if they are directly related to, and necessary to bring about, an M&A transaction. Such an agreement then constitutes what is known as an ‘ancillary restraint’ that falls outside the cartel prohibition. This is also set out in Article 10 of the Competition Act, the main condition being that the non-compete does not go beyond what is necessary, thus setting limits on the products/services that fall within the scope, the geographical reach and the duration of the non-compete.
The rationale behind exempting non-competes in the context of M&A transactions is that the buyer ought to have some protection from the seller in order to protect the value of the investment. With its knowledge of the market and customer base, the seller could otherwise effectively compete with the buyer immediately after the transfer, thereby effectively wiping out the buyer’s investment.
Limitation in duration
As regards the duration of non-competes, the basic principle is that – according to the European Commission – they are permitted for a period of two years in the event of transfer of goodwill. If know-how is transferred in addition to goodwill, the parties may agree on a three-year non-compete. A longer period may be permitted if it is justified in individual cases. However, the Netherlands Authority for Consumers & Markets (ACM) has also occasionally reduced a non-compete in the context of an M&A transaction in the IT sector to a period of one year because there was no justification for a longer period.
A non-compete for a longer period is not directly prohibited. A case in point is the 2017 Arnhem Leeuwarden Court of Appeal ruling on a non-compete clause on the sale of Cordial. The parties agreed that the seller was not permitted to compete with the buyer and the target for a period of five years. The Court of Appeal found that insufficient arguments had been presented to conclude that the non-compete violated the cartel prohibition.
Another case before the Arnhem-Leeuwarden Court of Appeal in which a five-year non-compete was also justified related to the sale of a company that produced and resold sorting systems. The Court of Appeal reasoned in this 2014 case that the market for sorting systems had a high level of customer loyalty and that the sorting systems lasted 15 years. Partly because the seller had all the specific knowledge of purchasing and sales channels at its disposal for many years, that sufficed for the Court of Appeal to conclude that this was an exceptional case. A five-year non-compete clause was therefore justified.
Limitation in scope
When formulating a non-compete, the target’s activities and catchment area must also be considered in order for it to qualify as an ‘ancillary restraint’. For this purpose, non-competes must be limited to those products and services that constitute the target’s economic activity. This may also include products and services that were at an advanced stage of development at the time of the transaction, as well as products and services that have been fully developed but have not yet been marketed.
In principle, the geographical reach of the non-compete must furthermore be limited to the target’s existing catchment area. Non-competes may also relate to those areas where the target – at the time of entry into the transaction – has concrete plans to expand its business and has also invested in that step.
Different rules for joint ventures
It is important to note that different rules apply to joint ventures that operate independently in the market (known as full function joint ventures). In order for the joint venture to be successful, parent companies may agree, throughout the life of the joint venture and with regard to the market in which the joint venture operates, that the parent companies and the joint venture will not compete with each other. But this applies only between the individual parent companies and the joint venture. Non-competes between the parent companies will not qualify as exempt ancillary restraints.
Risks if non-competes are too broad
Non-competes that are not directly related to the transaction or are not necessary are assessed on the basis of the cartel prohibition. Whether a clause is prohibited depends on the circumstances of the case, taking the economic and legal context into account, including the market position of the companies involved.
The moment the non-compete violates the cartel prohibition, it is void by operation of law. In that case the non-compete never applied and neither party may invoke it. In theory, the party that was wrongfully kept out of the market is also entitled to compensation. Moreover, the parties run the risk of a competition authority such as ACM imposing a (high) fine. So far, ACM has not imposed fines for this type of violation.
A case in point of the nullity of a non-compete is the 2017 judgment on the Thermagas sale. In that judgment, the Arnhem-Leeuwarden Court of Appeal ruled that, also in light of Thermagas’s market share (12%), the agreed non-compete for a period of five years after the sale was in breach of the cartel prohibition. According to the Court of Appeal, no grounds had been presented that justified such a duration.
Another example is the 2015 Rotterdam Court ruling on the sale of Wandflex. The court ruled in that case that the non-competes in the purchase agreement (for a period of ten years) and in the management agreement (for a period of five years) went beyond what was necessary for the sale. However, the court upheld the non-competes because it had been insufficiently substantiated that Wandflex’s market share exceeded 10%, which meant that it had not been established that the market share threshold of the de minimis provision (Article 7 of the Competition Act) had been exceeded.
Also in a case before the District Court of Midden-Nederland earlier this year, a non-compete was upheld because no evidence had been provided that the turnover and the market share thresholds under Article 7 of the Competition Act had been exceeded.
However, parties should be cautious when relying on the de minimis provision when agreeing on a non-compete, because the correct application of the de minimis provision based on the market share (<10%) requires a careful analysis of the product market and geographic market concerned. Moreover, no actual or potential interstate effect may be involved.
In light of the varying judgments on the admissibility of non-competes, the safest option for parties is to remain within the framework of the Commission Notice when formulating a non-compete. If they do exceed that framework, a sound justification of the need for the non-compete and a clear definition of the product scope and geographic reach may ultimately make all the difference.
Information on dawn raids by ACM and the European Commission can be found at invalacm.nl.
Follow Maverick Advocaten on LinkedIn
