2.1 Introduction

More than 20 percent of private homes in Norway have an alarm system (approx. 400,000 homes), which happens to be one of the highest penetration rates in the world.Footnote ¹ During the years 2008–2011, the private alarm market experienced a significant consolidation through a wave of mergers and acquisitions. The market structure became essentially a duopoly comprising companies Sector and Verisure with a joint market share of more than 80 percent that increased to over 90 percent, and a Herfindahl index of almost 4500 (at the national level) by 2019. While there is local geographic variation, Sector and Verisure are both present in most regions across Norway.

In June 2017 the Norwegian Competition Authority (NCA) did a dawn raid at the premises of Sector and Verisure. Essentially based on extensively documented direct communications between the two companies’ CEOs, the investigation revealed a collusive market-sharing agreement. The market sharing was related to door sales, which is a key marketing channel in the private alarm market. More specifically, the two companies had agreed to abstain from door sales to private homes marked with the rival company’s sign indicating it provided alarm services. Deviation from the agreement was detected by monitoring customer churn across the private alarm companies (and, in two instances, when a Verisure sales representative knocked on the door of the home of Sector’s CEO!).

The NCA found that Sector and Verisure had violated the prohibition against harmful collusive practices (§ 10) in the Norwegian competition law, which essentially is harmonized with EU competition law (Art. 101). For this violation, which is dated from August 2011 to the dawn raid in June 2017, the NCA imposed a combined record-high fine of NOK 1.2 billion (€ 120 million) on the two companies. There was no leniency notice or settlement procedure applied in this case. Five days after the Statement of Objection in 2019, Sector announced that it would pay its fine of NOK 467 million. Verisure first challenged the NCA’s decision but eventually paid the full fine of NOK 766 million after the Competition Appeal Tribunal upheld the decision in late 2021.

This cartel case offers a unique opportunity to understand how illegal collusive agreements are formed and sustained in practice. We know from economic theory that for a cartel to succeed, it has to overcome several challenges.Footnote ² Seen in retrospect, the collusive agreement between Sector and Verisure had important prerequisites for success. First, external instability was not a serious threat. Shortly before the cartel formed, the private alarm market had effectively turned into a duopoly by a series of acquisitions, and entry (or expansion) of potential competitors was deterred by an aggressive customer win-back strategy by both companies. Second, the market-sharing rule was simple and transparent. Door sales was the key sales channel to private homes. Homes fitted with an alarm system were marked with highly visible logo shields of the alarm company, which were meant to deter burglars. After the series of mergers prior to the start of the cartel, Verisure and Sector each had about half of the private homes with an alarm system under contract. By abstaining from knocking on doors marked with the rival company’s sign, the two companies implemented a simple and effective market allocation scheme. As far as we know, the two companies never coordinated on prices; they simply agreed not to poach each other’s customers using the door sales channel.

The remaining challenge for Sector and Verisure was making sure the collusive agreement was internally stable. Given a market-sharing arrangement, it is always tempting to deviate by stealing a customer from the rival. To understand how the two companies managed to ensure internal stability of their collusive agreement, we will examine in detail the extensive communications between the CEOs. We will also describe the various measures that Sector and Verisure took to monitor compliance and how they punished deviations from the market-sharing agreement. Interestingly, there are two episodes of collusion breakdown, or “war” (as it was referred to by the CEOs), during the cartel period, which helps us understand what can trigger a punishment phase and how companies manage to re-establish the collusive outcome.

This chapter is structured as follows. Section 2.2 describes the industry and how the cartel started. Section 2.3 presents briefly the core elements of internal stability for cartels in general, the challenges in this case, and how the two firms communicated. With that as background, Section 2.4 examines the communication, monitoring, and punishment observed in this cartel case in the context of the criteria for internal stability. Finally, Section 2.5 concludes with the lessons learned from this case study and a discussion on implications for competition policy in detecting collusive market-sharing agreements.

2.2 The Industry and How It All Started

The market for alarm systems in Norway has been growing sharply in the past few decades and achieved an annual turnover in 2021 of more than NOK 2.4 billion (USD 240 million) or NOK 480 (USD 48) per capita. The private and business segments of this market are quite different. Business customers require more complex equipment and services than private customers. Companies serving the business segment have higher costs, which makes them uncompetitive in the private segment. Thus, the alarm companies tend to specialize in one market segment, as is the case in the Norwegian market. Relevant to understanding this cartel, Sector and Verisure do not compete actively for business customers, and vice versa for the alarm companies in the business segment.

The private alarm market is subscription based, where customers pay (usually) a monthly fee for the alarm system and services offered by the alarm companies.Footnote ³ The package includes an installed (burglar and fire) alarm system that is connected to the company’s alarm center. Most of the companies, including Sector and Verisure, have their own emergency guard team. When switching providers, the customers need to have an alarm system installed by the new provider, as the alarm systems are not compatible across providers. For new customers, the companies typically offer the installation of the alarm system for free or at subsidized prices with a minimum period contract for their security services. Thus, the switching costs for customers are low.

The key sales channel to win new customers in the alarm market is door sales. Each company has a team of traveling sales agents who visit the premises of prospective customers and knock on doors. Phone sales are also used but mainly as a reactive sales channel for customers that check alternative offers or to convince customers that are considering switching providers to stay (which is referred to as win back). Advertising can also cause consumers to approach companies for an offer.

At the start of 2008, the private alarm market in Norway comprised four large companies – Sector, Verisure (named Securitas Direct until 2014), Hafslund, and G4S – and a range of small companies. However, during the ensuing years, Sector and Verisure gradually acquired a series of both small and large rivals. Table 2.1 summarizes the acquisitions by the two companies that were above the merger control thresholds and thus the NCA was notified.

Among the list of acquisitions in Table 2.1, two stand out. The first is the acquisition of Hafslund’s private alarm customer base by Verisure in 2008. This acquisition made Verisure by far the largest firm in the industry. Sector was the second largest, and G4S the third largest. In early 2011 the three companies had a joint market share of almost 80 percent – Verisure (43.6 percent), Sector (19.1 percent), and G4S (16.2 percent).Footnote ⁴

The second major acquisition is Sector’s acquisition of G4S’s private alarm customer base at the end of 2011. This acquisition essentially turned the market into a duopoly where Sector and Verisure had a joint market share of more than 80 percent with a competitive fringe of smaller rivals having the remaining 20 percent.Footnote ⁵ After the 2011 merger, the split of market shares between the two companies was fairly symmetric with Verisure having only a slightly larger share than Sector. By 2019 the joint market share of Verisure and Sector had increased to approximately 90 percent and become less symmetrically distributed, as Sector had 34 percent and Verisure had 56–57 percent. The third largest company was Nokas, with a market share of around 2 percent.Footnote ⁶

Figure 2.1 shows that both Sector and Verisure had strong growth in sales revenues (solid lines) from 2008 to 2017, with two significant jumps. The first jump is for Verisure from 2008 to 2009, which reflects the acquisition of Hafslund. This major acquisition almost doubled the annual turnover for Verisure. The other significant jump is for Sector from 2011 to 2012, doubling the company’s annual sales revenues by acquiring G4S. The sales revenues of both Sector and Verisure continued to grow in parallel, partly due to further acquisitions of smaller rivals by both companies. These mergers contributed to creating a market structure that was more conducive to collusion.

Figure 2.1 Turnover and EBITDA of Sector and Verisure, 2008–2017.

Figure 2.1 also displays the development in the two companies’ earnings (EBITDA), which are reported as a percentage of sales revenues. We see that their operating profits are increasing at a faster rate than their sales revenues. The increase in their earnings has been substantial for both companies over this period, as was the level of EBITDA of 30–35 percent at the end of the period. There is, however, an interesting break in this trend around 2012 and 2013, which coincides with the two periods where the collusive market-sharing agreement between Sector and Verisure breaks down (“war”) before it is re-established later in 2013.

How did the collusive market-sharing practice between Sector and Verisure start? In 2009, shortly after the acquisition of Hafslund, Verisure hired a CEO, who soon after his accession launched a new sales force called SWOT. This sales force was targeted with attracting customers from rivals through door sales. While Sector had used door sales as their key sales channel since the 1990s, the establishment of SWOT in 2009 marked a more proactive door sales strategy for Verisure. The CEO of Verisure said about the launch of SWOT:

According to the NCA’s decision, around the same time the two CEOs were in contact for the first time:

As we will describe later, this resulted in frequent, long-lasting communications between the CEOs of the two companies, involving email, phone, and even personal meetings in the coming years. Verisure’s competitive move to develop more door sales contributed to cartel formation between the two companies.

Sector’s new CEO from 2007 had previously worked at Hydro. According to Verisure CEO’s later testimony, he came from an industry where it was common to talk with competitors.Footnote ⁹ In October 2009 Sector’s CEO had sent an email to an employee of Verisure stating that “this is probably not in line with the industry’s attitude towards ‘business stealing.’”Footnote ¹⁰ This email was triggered by the fact that Verisure had started acquiring customers from Sector through door sales. After a meeting between the two CEOs in January 2010 (organized by the industry association for other legitimate purposes), Sector’s CEO wrote to the chairman of the company’s board: “I think we have the right contact in Direct [Verisure] now and we agree on the rules of the game. I will handle this.”Footnote ¹¹

After that, the two CEOs communicated frequently for years to come. They often did so through not only emails but also phone calls and face-to-face meetings. In Figure 2.2 we show a few of the important events in this industry, including the major acquisitions, periods of war, and some of the known meetings between the CEOs.

Figure 2.2 Timeline of documented key events and meetings between the CEOs of Sector and Verisure.

As indicated in Figure 2.2, and based on legal standard and evidence, the NCA defines the collusive market-sharing practice between Sector and Verisure to be established on August 17, 2011. The verdict refers to a fascinating start of the collusive conduct where on that day a salesperson from Verisure knocks on the door of Sector’s CEO, trying to sell him a Verisure alarm. Sector’s CEO immediately sends an email to Verisure’s CEO informing him that his house is well marked with Sector signs and asking why they sent their salespersons to Sector’s customers. Verisure’s CEO responds the same evening first by asking, “Did you buy?” But then he says they had an internal ban but cannot perfectly control their salesforce, and anyhow Sector started it all by capturing Verisure’s customers. However, he ends his email by conveying that they prefer to approach customers without an alarm system and not customers served by another company. The episode initiated a series of email exchanges between the two CEOs, which, according to the NCA’s decision, established the collusive market-sharing agreement between Sector and Verisure that lasted until the dawn raid in 2017.

2.3 Collusive Practice and Internal Stability

To sustain a collusive outcome, firms must agree on the specifics of one, achieve internal and external stability, and avoid detection by the competition authority. While obviously the collusive agreement between Sector and Verisure eventually failed as regards the last condition, it lasted several years.

The alarm market has features that make it easy to coordinate on a collusive outcome. Specifically, the companies are actively approaching customers, and the allocation of customers is public information for it is indicated by an external marker at a customer’s home. In contrast to other retail markets, where consumers approach suppliers, there is potential for a no-poaching agreement simply by agreeing to not approach the doors of other firm’s customers.

External stability was served by the pre-cartel acquisitions of many smaller rivals and the win-back strategy of Sector and Verisure. Since the market for alarm services is subscription based, the original provider would be notified if a house owner decides to switch to another provider, as this involves a termination of contract. This notification facilitates an effective monitoring of customer churn across providers. The providers are thus informed about the identity of the new supplier and can, before the subscription is canceled, give a matching offer to the customer.Footnote ¹² By having such a win-back strategy, they will, in many cases, have prevented a new supplier from capturing their customers. This aggressive strategy toward rival firms explains why the two companies did not lose market share after the cartel was established. In fact, their joint market share increased from 80 to 90 percent during the years they colluded.

The key challenge for Sector and Verisure, however, was to ensure the internal stability of the collusive agreement. The colluding firms had adopted an arrangement that incentivized compliance with the agreed-upon no-poaching outcome. This involved two key elements. First, they had a punishment for deviating from the collusive agreement and this punishment must be severe enough that the expected profits from complying with the agreement are higher than the expected profits from deviating. Second, deviations were observable to the collusive firms, so that punishment can be implemented quickly, thereby ensuring the expected profit from cheating remained low.

Monitoring is a key element of most collusive schemes as it facilitates the detection of noncompliance by a cartel member. Helpful for monitoring, the private alarm market in Norway is transparent along several dimensions, as pointed out in the NCA’s decision and by the parties themselves. When a private house owner signs a contract with one of the alarm companies, the firm installs the alarm system and puts a sign on the house. The sign states explicitly that an alarm is installed in the house and names the company that provides the alarm system. Obviously, the main reason for marking the house with a sign is to deter potential burglars from breaking into the house. But, as we will see, the sign can also be used as a signal to a rival firm’s sales agents to abstain from knocking on the door and move on to another house. In other words, the alarm sign on a house can be an effective coordination mechanism.

Verisure and Sector used the monitoring information of customer switches to calculate the customer churn (usually at monthly level) for their closest rivals. When a customer filed a cancelation, a sales person would contact this customer (usually by phone) before the cancelation was effective. During this conversation the sales person would ask why the customer canceled and what they could do to convince the customer to stay, such as matching a rival’s offer or lowering the price. From this communication, the two companies also had a precise record of how many customers they took from their rivals. The companies also knew that the rival was monitoring customer churn. The CEO of Verisure explained it this way:

In many markets, including the private alarm market, prices are not directly observable to rival firms, as they may be personalized or include individual rebates. In such cases, monitoring is made easier if the collusive outcome is determined as a market allocation scheme, for example assigning specific customers to each cartel member as in this particular case. Since the rivals are very small, any customer switching is very likely to be between Sector and Verisure, so each firm can assume that the loss of sales is very likely due to cheating by the other firm. But even in this case there is no guarantee that cheating took place, since it could be that some customers are leaving the market or other customers are entering the market.

With deviation effectively monitored, cartel stability also requires some prospects for a punishment taking place that would deter the firms from deviating in the first place. To make such a threat effective, several conditions should be met. First, the punishment must be simple and easy to understand, so the cartel members will know exactly what to expect and can easily calculate the consequences of cheating. Second, it should be a credible threat so each firm can expect the punishment will be imposed. Third, the punishment must be tough enough to make deviation unprofitable and at the same time designed such that it is possible to return to a collusive outcome.

While the collusive agreement between Sector and Verisure was clear and simple (i.e., do not knock on doors of the rival’s customers), there were some challenges with the interpretation of the customer churn data. Sales of alarm systems could be done in other ways than through door sales. It could be by phone sales or by trade fair sales, implying that there could be some customer churn, even if they complied with not knocking on the doors of the other company’s customers. Thus, the churn numbers had to be interpreted with some care in terms of whether they truly reflected deviations from the market-sharing agreement.Footnote ¹⁴

There was also a potential agency problem for the companies, as the remuneration of the sales agents was partly based on their sales volume. Thus, sales agents had an incentive to knock on all doors, including those of the rival. This could generate unintended customer churn because of lack of compliance by each company’s sales force. It was particularly Verisure that had its sales force on such contracts. To deal with this challenge, the company monitored its sales agents and sent clear instructions to its sales force to avoid Sector customers. There are several telling examples of such internal communications.

2.4 Cartel Communications and Conduct

So far we have explained that the collusion involved a simple coordination device which was supported by detailed monitoring. But to illustrate how they managed to punish deviations, we need to look into the communications between the two firm’s CEOs. Let us first describe those communications, and then discuss the cartel’s strategy concerning deviations and punishments.

2.4.2 Conduct: Deviations and Punishments

This case illustrates the challenges cartels face in sustaining their agreements, even in a setting with two firms, a rather simple and transparent coordination mechanism, and the absence of any serious threats from non-cartel members.

That they started coordinating their behavior around 2011 makes sense given acquisitions resulted in those two firms controlling almost 80 percent of the market for alarm systems for residential homes. And a win-back strategy led to a quite tough response to those outsiders trying to capture their consumers and thereby stopped them from growing and being a threat to the cartel. A system of regular communication between the two CEOs was established already right after the new Verisure CEO arrived in 2009.Footnote ¹⁶ In 2010 there were instances where they asked whether the rivals’ business stealing is a misunderstanding, for example caused by an overly ambitious sales agent, or was instead part of a new business strategy. In early 2011 we then see the first reference to a tit-for-tat strategy concerning existing customers, combined with an encouragement to go after only new consumers (no business stealing).

According to the NCA’s decision, the cartel started after the CEOs explicitly informed each other, on August 17, 2011, that they had instructed their sellers to abstain from knocking on rival doors. This market-sharing mechanism could easily be implemented, since a customer with an alarm has signs on their house with the name of the alarm company. Note, though, that there is really no clear date on when the cartel started. The two CEOs had increasingly engaged well before, while communicating their strategy to each other and signaling a market-sharing mechanism – a common understanding of not knocking on each other’s customers’ doors.

Even in such a transparent market, there is still some uncertainty about what is going on. As far as we see, deviations arose from two different sources. First, there is constantly a question whether sales personnel chose to deviate on their own. Approaching the rival’s customers could enable them to meet their sales targets. Second, after the acquisition of a small rival, deviations could emerge because of uncertainty as to whether those new customers for the acquiring firm are part of the no-poaching agreement.

The deviations perpetrated by sales personnel are illustrated by some of the communications between the CEOs as, for example, when one asks the other whether there is a new campaign initiated by the rival or just a sales agent operating on their own. At play here is a fundamental principal–agent problem, where incentives for each sales agent to acquire new customers can make it tempting to knock on a house with the rival’s sign. It also indicates that they are aware that cheating can take place and that each has the incentive to not inform the other firm about cheating, and that they have to work to verify the reason for a customer switching providers.

The customer churn between Sector and Verisure varied a lot during the cartel period. This is illustrated in Figure 2.3 which displays the monthly customer churn from Sector to Verisure (and a few smaller rivals) over 2009–2017 normalized at the churn level in the first month of observation (which is January 2009).Footnote ¹⁷

Figure 2.3 Monthly customer churn from Sector to Verisure and smaller rivals, 2009–2017 (divided by the churn level in January 2009).

Source: The Norwegian Competition Authority

Figure 2.3 shows a striking pattern. The customer churn for Sector is stable at a low level until spring 2012 when it sharply increases and peaks by a factor of more than 50 by the fall of 2012. The monthly churn then drops significantly during spring 2013 before it peaks again in fall 2013 by a factor of almost 80 by fall 2013. These changes are massive.

Interestingly, the sharp increases in customer churn from Sector to Verisure and smaller rivals correspond precisely with the two war periods between the companies, which we described in detail above.

The first war broke out in February–March 2012. This illustrates our second source of deviation, the acquisition of a small rival. This war was triggered by Sector’s acquisitions of the private alarm customer bases of several rivals, including GS4, Nokas, and BKK. As a response to these acquisitions, Verisure started a campaign to attract customers of the companies that Sector had acquired. Prior to the acquisition, Verisure could have competed for those customers, but now those customers were supplied by Sector and, from Sector’s perspective, were not up for grabs. In essence, Sector’s acquisitions had caused there to be a disagreement between Sector and Verisure about the market allocation scheme.

Given Sector considered Verisure’s campaign to be a hostile maneuver that violated their market-sharing agreement, it chose to respond aggressively by going after Verisure’s customers. To signal that this was retaliation (or punishment) and not a deviation, Sector sent a fax to Verisure with all the customers they had stolen from Verisure in the previous week. If it had instead been a deviation from the collusive outcome, as opposed to a punishment as part of the collusive arrangement, Sector would have kept that information hidden from Verisure.

An interesting observation is that, during the punishment phase, Sector and Verisure did not follow a tit-for-tat strategy but instead switched to an all-out attack on its rival. During the summer of 2012, when the first war escalated, there was no more “an eye for an eye” between the two companies. And the war lasted longer than initially planned. Verisure’s original plan was to fight until the end of July, but the war continued for several more weeks. Verisure also blocked all communication with Sector which, according to Verisure, created uncertainty about the toughness of the punishment and how long it would last. Although we do not know what determined the length of Verisure’s aggression, one possible explanation is that their plan was to acquire a certain number of customers before approaching Sector to end the war.

During the first war, it seems that Verisure fought more intensely than Sector, as illustrated by the fact that Sector’s CEO was the one trying to ask for “good weather and grace.” However, he was not able to get in touch with Verisure’s CEO by email or phone. Eventually they had a face-to-face meeting right before that the war ended, and it seems that Verisure had captured more customers from Sector than Sector had acquired from Verisure. Although we do not know what the CEOs talked about during that meeting, it appears that a physical meeting or at least close communication was needed to return to the tit-for-tat strategy they had before the war.

The second war was in the first half of 2013 and seems to be the result of a gradual escalation, where the tit-for-tat strategy led them to gradually steal more customers from each other. At some point, this led to an all-out war, where both firms launched campaigns to steal customers from each other. As in the first war, the approach was not tit-for-tat, but rather a punishment where each of them was apparently trying to grab as many customers from their rival as they could. One interesting distinction from the first war seems to be that Sector was more aggressive. Verisure noticed this, and there were internal discussions in Verisure on whether they would gain from ending the war. Right after that, the two CEOs met and the war ended.

One can interpret this conduct as consistent with an extended version of the classic tit-for-tat strategy or a combination of tit-for-tat strategy and a stick-and-carrot strategy with a tough punishment phase.Footnote ¹⁸

First, they took into account unintentional errors, as not every deviation brought a retaliatory response. Communication was critical in several instances, in particular when one CEO asks the other about the reason for some business stealing. It can facilitate collusion to be able to identify the deviations that are, in fact, a mistake and thus do not warrant an aggressive response.

Second, deviations could vary a lot in terms of how many customers are taken from the other company. There was always some business stealing going on in this market, even after the companies both had declared that the war is over. It seems as if they strived to keep business stealing to a minimum, while recognizing that it would not entirely disappear.

Third, there was a systematic departure from the tit-for-tat strategy when the companies entered into a war with a punishment tougher than tit-for-tat. One interpretation is that the punishment is as severe as it can be, and then the question is how long will it last. In the first war, Sector was the one that had to back down, while in the second war it seems that Verisure had to give in. In this respect, it was more of a stick-and-carrot strategy with a tough punishment phase, rather than a tit-for-tat strategy.

Fourth, direct communication seems to have been crucial for this cartel to succeed. The mere scope of the communication – through emails, phone calls and face-to-face meetings – reveals their importance for internal stability. In particular, the evidence shows how reversion to the no-poaching outcome can be difficult to do just by signaling through market behavior. Indeed, a meeting where the CEOs met in person was apparently the turning point both in fall 2012 and fall 2013 when they ended the two wars.

Finally, the two wars they fought seem to have been very important for the subsequent coordinated behavior. From internal emails we see that they referred to earlier wars, and told each other that such an outcome should be avoided. Less frequent direct communications are observed after the two wars, indicating that they had a better understanding of each other’s strategy. The price structure they both used – providing free alarm installation to new customers – reduced the switching costs new customers would face. It is then no surprise that they refer to earlier wars, since low switching costs can lead to tough competition when they fight for customers. The prospect of intense competition in response to evidence of noncompliance can deter deviations and support a stable collusive outcome.

2.5 Lessons to Be Learned

From the perspective of competition law enforcement, it is relevant to ask what can be learned from this particular cartel. It is well known that having fewer firms in an industry may increase the risk for (tacit or explicit) collusion and thus lead to a larger scope for a cartel to succeed. As a result, merger control can and should play an important role in industries with a potential for collusion.Footnote ¹⁹ This case supports that point, though in an unfortunate way. The two cartel members made numerous acquisitions that were notified to the NCA, and at least two of these acquisitions were decisive for the industry to be dominated by those two firms. Although we have no indication that the mergers were part of a plan to form a cartel, still the success of the cartel in this industry was at least partly due to the NCA allowing a market structure to form that facilitated collusion.

Another lesson is that the market share of a cartel that is not industry wide can increase, even though independent fringe competitors can benefit from their umbrella effect to raise prices. In this case, the joint market share of the colluding firms increased (from 80 to 90 percent) during the cartel period. This is usually not consistent with standard theory of explicit collusion when the cartel does not involve all firms in the market.Footnote ²⁰ In that case, non-colluding firms would usually benefit from the cartel by increased market share and higher margins. However, in this case the collusive agreement between Sector and Verisure applied only to customers of the two companies. Competition between them and other small suppliers for customers of rivals (or customers without an alarm system) remained, and indeed both companies competed aggressively to steal customers from non-colluding companies. This explains the rising joint market share of Sector and Verisure.

Another lesson is that explicit coordination may be more often necessary than sometimes thought. Considering the industry characteristics in the Norwegian private alarm market, there is a large potential for coordinated behavior: just two large firms, great transparency of sales, and a simple market-sharing device. Such a setting should be expected suitable for (lawful) tacit collusion rather than (unlawful) explicit collusion (see, for example, Fonseca and Norman, Reference Fonseca and Normann2012). But that is not what happened in the Sector–Verisure case. Coordination in the real world might require express communication even in a setting with considerable potential for tacit understanding. Note also that there is, as far as we know, no communication between the parties about prices, only a no-poaching arrangement.

Let us think about some collusive markers suggested by the Sector–Verisure case that could be used for other markets. In subscription-based markets, like the alarm market, customer churn can be an informative indicator for a possible collusive agreement, at least when the market ticks off the boxes in terms of being suitable for collusion. In such circumstances, competition authorities may look for patterns in customer churn similar to the ones in this case. In particular, if one observes periods with high and unstable churn being followed by periods with low and stable churn, this may suggest that a collusive arrangement has been implemented by the companies. Usually, such data are not publicly available, which means that competition authorities would need to collect this information either from third parties or from the parties themselves. Still, this could be a useful ex officio strategy.

Another insight from this case study is that direct communication can be extremely important in order to implement and sustain a collusive agreement. When monitoring is imperfect – as churn is not solely due to business stealing but may have other sources like opportunistic sales agents, customers entering or exiting the market, or simply natural variations – then communication plays an important role. Indeed, communication can be used to signal that you see the rival engaging in a deviation and then putting forward threats of retaliations, but also clarifying misunderstanding that may avoid a mistaken and very costly war. Equally important, communication might serve as a tool to re-establish the collusive agreement from a breakdown due to deviations and retaliations. Given this important role of communication, it is likely that competition authorities may find evidence of such, though more sophisticated ways of communication may make it difficult to detect in other cases.

This case also shows that once a cartel is established, there might be less need for direct communication. In this particular case we observe that after two wars the companies both realized what might happen if they deviated, and direct communication happened less frequently. This observation points to how difficult it can be to shut down a cartel once it has achieved internal stability. Cartel detection and enforcement through high fines might not be enough, since direct communication might no longer be needed to maintain the agreement. This calls for other means, for example director disqualifications and imprisonment to try to change the culture in the corporation and thereby induce competitive conduct. Another alternative could be structural remedies, such as a sale of assets, which is already in use in merger control in many countries and a few monopolization cases in the USA. As we observed in this case, acquisitions preceded cartel formation and it may take the sale of assets to destabilize the cartel.Footnote ²¹

By way of conclusion, we would like to mention that we also believe this case study provides insights for future research on collusive strategies. Indeed, the switch from a tit-for-tat strategy to one more similar to a stick-and-carrot strategy was observed in this case. To re-establish the collusive agreement, the alarm companies implemented an even harsher punishment than tit-for-tat. As conveyed by one of the CEOs: “now you are asking for a real war.”