This summer the European commission (finally) announced it will start a formal investigation against Microsoft following Slack’s complaint concerning the (abusive) tying or bundling or Teams to the Microsoft and Office 365 suites. Not long after, Microsoft came out with an official statement concerning the changes in its pricing and distribution strategy of Teams it will introduce in order to alleviate the concerns identified by the Commission. This week’s post discusses how, if pursued, the case may prove to be yet another landmark case on tying and bundling.
Factual background
During the summer of 2020, news came out about a claim made by Slack against Microsoft for abusing its dominance by tying Teams to the Microsoft and Office 365 software packages. The claim at the time was that Microsoft had been forcing millions of users to install Teams and denied them the possibility to uninstall such software. Furthermore, claims were also made about the fact that Microsoft was not being transparent enough about the usage fees of this package deal nor about the fees involved in using Teams. The addition of Teams to Office 365 put Slack at a disadvantage when it came to competing for enterprise customers interested in productivity and connectivity software. Unlike Microsoft, which offers an entire collection of productivity and connectivity tools in its Microsoft and Office 365 suites, Slack operates more like as a modular communication platform that can be customized based on customer preferences by linking various third-party software solutions together with Slack’s own communication services.
Accordingly, instead of providing an entirely integrated productivity and connectivity software suite, Slack provides an enterprise communication service that can be used as a hub to connect various third-party service apps such as Zoom, Gmail, Dropbox and many others to achieve a similar goal while offering customers more customization options. In essence, Slack offers its customers the ability to create their own suit out of different software components offered by different providers as an alternative to the various providers of full native suits. Seeing as the communication software is the heart of Slack’s business model its objection to Microsoft’s new strategy is understandable. Getting Microsoft and Office 365 business and enterprise customers to use Teams as their default communication solution undermines Slack’s modular solution since it essentially cuts out the hub of the entire system Slack has been trying to build. Without the need to consider utilizing third-party communication platforms there is then no opportunity for Slack to offer its core service or an alternative type of solution for connectivity and productivity suites. Consequently, this, in turn, may also affect many of the third-party solutions that Slack allows to interconnect which would otherwise benefit from the ability to form part of customized suits thereby increasing their chance of survival as standalone products or services.
Despite the evident competitive concerns, the undeniable resemblance with the previous Microsoft cases and the enforcement momentum against big tech, the case seemed to fade off quite quickly after making the headlines at the time. Now, approximately three years later, the European Commission decided to step in and launch an official investigation into the matter. Despite this delay, if pursued, the case may turn out to be yet another landmark case on tying given the number of challenging aspects it may have to address.
Market definition – finding dominance and framing the abuse
As with any abuse of dominance case, the first difficult hurdle that needs to be overcome is the definition of the relevant market, which is a legal pre-requisite for finding dominance (see Case T-62/98 Volkswagen v Commission, para 230; Case C‑7/97 Oscar Bronner, para 32; Case C-52/07 Kanal 5 and TV 4, para 19). In this case, where the type of abuse alleged is tying or bundling, the market definition holds another important function and that is to establish the existence of two or more separate products or services which have been tied or bundled together. If the market definition does not support such a conclusion, there can be no finding of such abusive practices. Evidently, however, the market definition in such a case cannot be done differently for these two purposes.
Accordingly, there is much on the line when it comes to this element in the case against Microsoft. In practice, the most likely outcome is a definition of two relevant markets. One for productivity software suites composed of word processing, presentation tools, spreadsheet creator and potentially an email and agenda manager (i.e., in the case of Microsoft: Word, Powerpoint, Excel and Outlook) and a second relevant market for (video) communication and collaboration tools (i.e., Microsoft Teams). Such a market definition would be suitable for a case of tying where the traditional core Office 365 or Microsoft 365 package would constitute the tying product, for which dominance must be established, and Teams would entail the tied product. Alternatively, this could also support a (pure) bundling case if it can be shown that the two components (Teams and the rest of the suit) could not have been obtained on a standalone basis. Although it is not entirely clear from the wording of the press release which of the two abuses would be argued, it does appear that the market definition in the case will follow these lines.
This approach may seem, on the face of it, rather arbitrary. After all, why should video communication software be separated from productivity software when it has become an indispensable tool for hybrid and remote working schemes? While this is to some extent true, treating these software solutions as one integrated product may sidestep the different characteristics of productivity versus communication tools which serve different purposes and meet different needs. These differences are, to some extent, also acknowledged by the Commission’s practice in the context of the EUMR where these categories of products have been overall addressed separately, albeit not in an extensive or exhaustive manner (for example, see Microsoft/LinkedIn, paras 19-28).
Furthermore, the division between these two categories of tools is, to a great extent, also displayed by the current market alternatives to Microsoft’s offer. Alternatives for Microsoft Office tend to come in 3-in-1 or 4-in-1 packages consisting of word processing software, presentation tools, spreadsheet creators and at times also email and agenda managers (e.g., LibreOffice, WPS Office and many others). By contrast, (video) communication and collaboration software is often offered as a stand-alone product (e.g., Zoom or Slack). Such examples would be, according to the case law, on tying sufficient to shoulder the existence of separate products. In fact, this goes even beyond what is required as it is proof of separate demand for both the tying and tied products (Commission Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty to abusive exclusionary conduct by dominant undertakings, para 51).
In this respect, Google may be the only other market player (in the EU) that is able to offer a comparable suit to that of Microsoft. This alone, however, cannot suffice as evidence that the competition in the market of productivity software solutions has evolved into competition among bundles or packages that include both productivity and (video) communication and collaboration software. This is not to say the market will not evolve in such a manner in the future. However, currently, this does not seem to be the state of play. Moreover, and perhaps most importantly, this was even less likely the case when the complaint was initially made in early 2020, i.e., at the pre-pandemic stage, when (video) communication and collaboration software was (far) less of a necessity in the professional working sphere than it is today. Accordingly, even if the market were to evolve into competition among multiproduct suits that include (video) communication and collaboration software, that does not take away from the fact that this was not the case in the past nor that such practices were potentially abusive in that respective timeframe given the ex-post nature of Article 102 TFEU.
Anti-competitive concerns and effects
When it comes to the anti-competitive concerns raised by tying and/or bundling practices, economic literature indicates risks of (i) foreclosure in the tying and/or tied product markets; (ii) deterrence of entrance in the tying and/or tied product markets as well as a (theoretical) third product market for a novel product capable of replacing the combination of the tying and tied product; (iii) the extraction of supra-competitive prices in both tying and/or tied product markets. The manifestation of such potential harms in practice depends, however, on the circumstances of each case and the market conditions present at the time of the analysis.
When looking at the Commission’s press release, it is possible to identify a foreclosure concern on the tied market (the market for communication and collaboration products) and, to some degree, deterrence on the tying product market (the market for productivity software). In the particular case of Slack, one could even argue a potential foreclosure or prevention of entry to a third market for the product intended to replace the combination of the tying and tied product by offering a collaboration and communication tool that allows to create a customized software suite.
Bringing a case under Article 102 TFEU requires, however, more than theoretical anti-competitive concerns. While such concerns may suffice to open an investigation, reaching a finding of abuse requires a more in-depth analysis and proof of such potential effects. This is where the case if pursued, could become quite challenging, as it differs significantly from the previous Microsoft cases. The reason behind this is the nature of competition between the products involved in this case, namely collaboration and communication software.
In the Microsoft media player case, the tying undermined competition between the various available media players at the time. In this regard, the General Court and Commission found that the tying practices eliminated the incentives of consumers to search for competing alternatives to Microsoft’s WMP since it was offered to them for free and provided them with the same basic functionalities at an acceptable level of quality (Microsoft, paras 847-848; and Case T-201/04 Microsoft v Commission, para 1041). Since most consumers would indeed be interested in having a media player that works with the most common video and audio forms, it is easy to see how Microsoft’s tying would create a foreclosure effect in the market for digital media players. Although this is less explored with respect to the tying of Internet Explorer, it is not hard to see how, at the time, the same logic would apply in such context as well (para 39-54).[1]
When looking at Slack’s current dispute with Microsoft, however, the situation is clearly more complex as it concerns a more sophisticated product AND customer. Accordingly, it is not clear that in this case the mere pre-installation or addition of Teams to the Office suite would suffice to generate a similar anti-competitive effect as in the previous Microsoft cases. To start off, only the most basic version of Teams is entirely free of charge. This is also true with respect to Slack’s most basic version. In this segment of the market, there may indeed be a potential foreclosure effect in the tied product market as the pre-installation of the most basic version may reduce the incentive of customers interested in such version from looking for alternatives as in the previous Microsoft cases. One may wonder, however, how many of these customers truly exist given the fact that both Teams and Slack are essentially targeted at enterprise and business users. Given the limitations of the basic versions of such software, it is quite likely that most -if not all- users will typically opt for the paid versions of these tools. Here, the effect of pre-installation may not be as strong as with the free version. Committing to a licensing scheme and the adoption of a paid multi-user subscription for at least a year would require more than pre-installation. In this regard, the pre-installed free version offers perhaps a shortcut to giving potential customers the first real-life ‘pitch’ of the Teams software.
The above should not be interpreted as an indication that Microsoft’s practices pose no anti-competitive concerns, as it clearly provides Microsoft with an advantage over competitors, otherwise, it would not make sense to implement it. Nevertheless, this case should also not be treated as yet another Microsoft sequel. Proof of the alleged anti-competitive effects of Microsoft’s practice will be less evident than in past cases under such circumstances. Consequently, if pursued, this case will put the effects-based approach to tying cases to the test as establishing a finding of abuse will require a more in-depth analysis. Furthermore, if the anti-competitive effects of such practices are rather limited, this case may also trigger once more the discussion on the matter of de minimis in the context of Article 102 TFEU. Although the concept of de minimis in Article 102 TFEU case has been rejected several times, it would not be hard to see how this case could give rise to this discussion again, as a potentially limited anti-competitive effect would serve as the foundation of a (multi) million -if not billion- euro fine. Admittedly, from a formal legal perspective, there is not much standing in the way of such an outcome, which would arguably also contribute (significantly) to deterrence. Nevertheless, one can imagine how a potential friction with the principle of proportionality would begin to surface.
The complexities involving the potential anti-competitive effects, in this case, will arise solely in the circumstances depicted above, namely in the case that the pre-installed version of Teams was truly the most basic (and free) version of this software. Gaining access to the more advanced (enterprise) functions of Teams would therefore require upgrading to the paid version(s) of Teams services. If, however, the Teams version bundled with Microsoft 365 and Office 365 was an upgraded version from the outset, the potential anti-competitive effects could be, perhaps, even more significant than in the previous Microsoft cases. This is because such service would have been paid for already by the respective customers (whether they were made aware of this or not) and the fact that any other alternative would also be a paid (upgraded) service. In such circumstances, the incentives of customers to look for competing alternatives would be significantly diminished if not completely eliminated. Accordingly, in such a scenario the current investigation would very likely turn out to be almost identical with the previous cases.
Efficiency arguments
As with any case of abuse, the case of Microsoft may also offer some room for an efficiency justification depending on which of the scenarios highlighted above actually took place. The most room for such a plea would evidently be in the situation where the pre-installation of Teams concerned the basic free version.
To start with, tying and bundling practices have been found capable of generating various efficiencies (as put forward by O’Donoghue and Padilla, at 599-602; and Jenkins and Kavanagh, at 205-208). Common efficiencies achieved by such practices are reduced search and transaction costs, reduced distribution costs and quality assurance. In this case, some of these can be said to be achieved to some extent.
As with many cases in the past, the success of such an efficiency defence will greatly depend on the indispensability of the concerned practices for achieving such benefits and the degree of harm that it caused to competition as a result. The pre-installation of Teams in the Office suite can, theoretically, achieve most of the common efficiencies associated with tying and bundling practices. Whether such a technical combination is indeed indispensable is a difficult matter. On the one hand, it could be argued that pre-installation goes too far and that a download link or prompt, for example, would be less far-reaching and lead to similar results. On the other hand, not offering pre-installation would reduce (at least part) the efficiencies that could have been achieved by this practice. The same applies when it comes to consider the degree of competition which may be harmed. On the one hand, the practices by Microsoft may significantly harm or even eliminate competition when it comes to the free basic version of Teams. On the other hand, the harm to competition in the markets identified by the Commission caused by this behaviour as a whole may be (far) less pronounced for the reasons discussed above.
Overall, the case practice of the Commission and the case law of EU courts do not offer much hope when it comes to the success of such efficiency defences, and mostly for good reasons. The combination of indispensable restrictions of competition which are not liable to eliminate competition when rolled out by dominant undertakings is very difficult to find in practice. To some extent, it can be argued it goes against the logic of this provision. A justifiable abuse of dominance seems simply contradictory. Consequently, it is not surprising that there has been no successful use of this justification at the EU level so far. Nevertheless, from all the major cases in the past few years, this one against Microsoft could be the one that has the most going for it. Perhaps not enough to successfully make use of this justification possibility but certainly enough to re-engage in the discussion of this option for justification so as to further clarify what a successful use of it would require.
In the event that the tying /bundling in this case concerned the upgraded version of Teams, the above-mentioned mitigating arguments lose their strength as it is hard to argue why a fully upgraded version of Teams must be offered in tandem with Microsoft 365 and Office 365. In such a case, the option of pre-installing the free version and constructing further upgrades as (above cost) mixed bundling is always the less harmful option capable of delivering similar efficiencies.
The implemented changes by Microsoft and further need for investigation
The complexity of the case and great stakes have not been missed by Microsoft, which on 31 August, published an official statement on its intentions to make several changes to its distribution strategy of Teams so as to remove the concerns identified by the Commission. The question is if that really is the case, or are there still good reasons for moving on with the case?
In its statement, Microsoft indicates that it would (i) offer an unbundled version of Microsoft 365 and Office 365 at a reduced price of 24 or 60 euro per year (depending on customer type); (ii) increase the interoperability of Microsoft 365 and Office 365 to work better with third-party solutions; (iii) create a new mechanism to enable third-party solutions to host Office web applications.
Although the nature of the bundles was not entirely clear so far, these changes made by Microsoft seem to hint at the scenario that the Teams version that was bundled with Microsoft 365 and Office 365 was indeed the upgraded version and not the basic free one. Consequently, the degree of potential anti-competitive harm involved in this case is more likely to be on the higher end of the spectrum. Therefore, if the Commission were to consider ceasing the investigation in light of Microsoft’s recent changes, it is very important that such changes are adequately scrutinized beforehand given the mishaps of the past in the previous Microsoft cases (as documented by Gore and van Rooijen).
The unbundling of Teams from Microsoft 365 and Office 365 is, in essence, the most evident step forward towards bringing, at least, some of the competitive concerns with regard to this behaviour to an end. This change would, in principle, only apply to new (potential) customers. Customers which have already committed to the bundle are not affected by this change until their contract is up for renewal. With regard to new customers, the effect of the suggested changes will depend on the significance of the price reduction and its potential to be considered predatory. The impact of this change on competition depends on how feasible it is for competitors to profitably offer alternatives to Teams for EUR 24 or less per year, which is the difference between the bundled and unbundled versions of Business versions of Microsoft 365 and Office 365. Similarly, the same question applies to the new enterprise suits which are sold only unbundled with the option to opt to Teams for an additional fee of EUR 60 per year.
If no such alternative options exist because the costs of offering such alternatives are higher, then the reductions and/or new fees introduced by Microsoft for Teams are highly unlikely to lead to any change in the market. In such an instance, it is then imperative to assess whether Microsoft itself is able to provide Teams profitably at such a price. If not, then such changes may have merely modified the strategy from a traditional tying or bundling into a combination of predatory pricing and predatory mixed bundling, which can be equally harmful to competition (as analysed by Nalebuff in 2003). If, however, Microsoft is able to provide Teams profitably for these new fees, the behaviour may indeed no longer be considered abusive as it has been established time and time again, EU competition law and Article 102 TFEU specifically are not concerned with protecting less efficient competitors.
Nevertheless, as pointed out, the pricing changes solely concern future customers and/or existing customers who consider renewing their contracts with Microsoft. Consequently, any market power that has been gained by Microsoft through this alleged abusive behaviour will not be impacted at the very least for the duration of its existing contracts with current enterprise customers. This is to a great extent the same type of solution that was accepted and criticized in the case of Google Android where the untying practices solely applied to Android devices sold after the decision on the finding of abuse. In the case of Microsoft, the potential magnitude of this shortcoming cannot be overstated. The strategy applied throughout the entire duration of the COVID-19 pandemic where the use of software skyrocketed, meaning that if market power was to be (abusively) gained through such actions, a significant degree of this occurred over the past three years. Accordingly, any strategy changes and/or commitments made by Microsoft must have some restorative element built in order to constitute credible solutions for this ongoing investigation.
With the abovementioned in mind, it is difficult to see why the recent modifications by Microsoft should lead the Commission to cease its investigation. The implemented changes do little to nothing to address the alleged competitive harm caused in the past. Furthermore, it is far from evident that such changes in the price settings of Microsoft 365 and Office 365 will change the market dynamics that led to Slack’s complaint (assuming it is justified). The indications to improve third-party interoperability and compatibility are certainly welcome, however, these general commitments to do better are by no means guarantees for better outcomes in the future.
Therefore, if anything, the strategy changes introduced by Microsoft should be a reason to push the case forward as it shows some willingness for cooperation. If the evidence in the case can truly sustain a successful abuse of dominance case, the very least expected is to use such leverage to obtain more far-reaching changes and commitments. Such changes and commitments should better address both the past and future competitive concerns (and harm) caused by Microsoft’s strategy. To the extent that such steps can be made, ceasing the investigation could be something worth considering in light of the otherwise costly resources that such a case would involve. Admittedly, however, this would have some clear disadvantages for the creation of valuable legal precedents for both public and private (follow-on) enforcement.
Outlook
The recent investigation against Microsoft may turn out to be yet another landmark case on tying and bundling. Depending on the specificities of the tied/bundled versions of Teams, the case will present the Commission with either a very complex case to deal with when it comes to proving the effects of such practices or a rather straightforward copy of the previous Microsoft cases.
Regardless of the scenario involved, if the Commission were to consider ceasing its investigation in light of the recent changes made by Microsoft, it is imperative that such a decision is not taken without such changes being extensively scrutinized in terms of pricing and includes the restorative steps adopted by Microsoft; provided, of course, that (some) anti-competitive effect can indeed be identified. The ball is now in the Commission’s court once more, time will tell which move it will choose next.
[1] To a large extent, however, part of the consumer inertia following pre-installation in this case seemed to stem from a lack of (technical) knowledge concerning other browsers and / or the possibility of downloading them.