The 11th Amendment to the ARC and Germany’s New Competition Tool

The 11th Amendment to the ARC and Germany’s New Competition Tool

PDF-Version: Click here

Suggested Citation: Wagner-von Papp, DKartJ 2023, 22-33

On 5 April 2023, the German government, represented by Robert Habeck (Vice Chancellor and Minister for the Economy and Climate Protection, The Greens) and Marco Buschmann (Minister for Justice, Liberal Democrats (FDP)) unveiled the Government Bill (RegE in German) to reform the Act against Restraints of Competition (ARC) in a press conference [in German, but subtitles can be auto-translated]. In the press conference, Vice Chancellor Habeck called the reform the ‘biggest reform in decades’, ‘perhaps the biggest reform since Ludwig Erhard’. Even if one does not want to go as far as that, it is clear that the planned reforms are not just tinkering around the edges. Prof. Dr. Florian Wagner-von Papp explains the planned reform here. As a service to our readers we also publish a consolidated rough English translation (no official translation) of the government draft that can be downloaded here: GWB11 engl.

The content of the Government Bill can be summarized under the following three headings.

  1. Strengthening sector enquiries by introducing, in a Draft-§ 32f ARC (Act against Restraints of Competition = Gesetz gegen Wettbewerbsbeschränkungen, GWB),
    • the power for the Bundeskartellamt to order remedies including divestitures, or to accept commitments, where the sector enquiry finds a ‘substantial and persisting malfunctioning[1] of competition’ which may, but need not result from an infringement of competition law (similar to UK Market Investigation References in the case of adverse effects on competition or the EU plans for a New Competition Tool), and
    • the power for the Bundeskartellamt to impose on undertakings the obligation to notify all future concentrations where the sector enquiry finds that there are indications that such future concentrations may significantly impede effective competition, provided that the acquirer has a domestic turnover of more than €50m and the target has a domestic turnover of €500,000 in the last business year;[2]
  2. strengthening the power to disgorge economic benefits by adding two presumptions to the current text of § 34 ARC: first, the presumption that an infringement of competition law results in a gain to the infringer, and, second, the presumption of a minimum gain in the amount of one per cent of the relevant turnover; and
  3. providing for powers for the Bundeskartellamt to support the Commission in the enforcement of the DMA as well as introducing rules to enable and facilitate private enforcement of the DMA.

This blog post will first outline the Draft Bill of September 2022 (I.) before describing the fierce criticism the Draft Bill faced (II.) and the changes made in the Government Bill (III.). After giving my own two cents on the Government Bill (IV.), I briefly attempt a gaze into the crystal ball to assess the chances of the future fate of the Government Bill in Parliament (V.). A risky endeavour, no doubt: As everyone knows, predictions are difficult, especially about the future.

I. The Draft Bill of September 2022

Prior to the Government Bill, a Draft Bill for consultation [in German] had already been published in mid-September 2022 (below II.), and this, in turn, had been preceded by a ten-point competition policy agenda on 21 February 2022 [in German].

The Draft Bill that preceded the Government Bill already contained all the salient features that would later be included in the Government Bill.

1. The German-style New Competition Tool and sector enquiries

The central innovation in the Draft Bill was the new § 32f(3)–(7), introducing a German version of a New Competition Tool, inspired by the UK Part IV market investigation regime and the EU plans for a New Competition Tool (which, in turn, had been modelled on the UK regime).[3] The Explanatory Note argued that the current German regime of prohibiting anticompetitive horizontal or vertical agreements (§ 1 ARC), abuses of dominant positions (§ 19 ARC), abuses of relative or superior market power (§ 20 ARC), and abusive conduct of undertakings of paramount significance for competition across markets (§ 19a), as well as ex ante merger control (§§ 35 et seq. ARC) was insufficient to protect competition comprehensively. The Explanatory Note identified gaps, for example, for cases of tacit collusion (algorithmic or otherwise), possibly common or cross ownership, or cases of unilateral market power in the absence of discrete abuses. While merger control was meant to protect against market structures arising in which such coordinated or unilateral practices could impede competition, merger control cannot protect against internal growth, against the accumulation of market power in the hands of the remaining undertaking when competitors exit a market, against concentrations that preceded the introduction of merger control, against concentrations that should not have been cleared in the past, or against concentrations that were permitted by ministerial exemption.

On the other hand, the Explanatory Note to the Draft Bill also emphasized that internal growth in particular was generally to be seen in a more positive light, and that the new tool was not meant to stifle the incentive to succeed in competition. The sentiment of the 1945 Alcoa judgment that ‘[t]he successful competitor, having been urged to compete, must not be turned upon when he wins’[4] comes to mind, even if it is not quoted in the Explanatory Note.

In order to fill the gaps identified above, § 32f(3) of the Draft Bill was to enable the Bundeskartellamt to impose all necessary behavioural or structural remedies where, following a sector enquiry, it is established that a ‘substantial, continuing or repeated malfunctioning of competition exists on at least one market or across markets’. § 32f(4) of the Draft Bill gave the Bundeskartellamt the power to order undertakings to divest shares or assets, provided that it is to be expected that the divestment substantially reduces or eliminates the malfunctioning of competition, and further provided that no milder yet equally effective remedies are available.

In contrast to the subsequent Government Bill, the Draft Bill contained much less detail and fewer safeguards (see below III.). With a grain of salt, the operative part of § 32f(3) of the Draft Bill version could be reduced to the sentence: ‘Where a substantial, continuing or repeated malfunctioning of competition exists on at least one market or across markets, the Bundeskartellamt may impose all necessary remedies of a behavioural or structural nature on undertakings.’ The draft provision mentioned a non-exhaustive list of possible remedies. It did not, however, define the new term ‘malfunctioning of competition’, and did not even include a systematic list of examples or theories of harm. Instead, it listed (in § 32f(5) of the Draft Bill) a non-exhaustive number of factors which were to be ‘considered’ when determining a malfunctioning of competition. These included, but were not limited to, the number, size, financial power, and turnover of the undertakings on the relevant market or markets, the market shares and market concentration, cross-shareholdings, market entry or exit barriers, and market outcomes.

Minister Marco Buschmann (Justice) and Robert Habeck (Economic Affairs and Climate Action) with a spokesperson from the Ministry (right to left), presenting the government bill of #GWB11.

In terms of procedural safeguards, § 32f(3) of the Draft Bill contained none, except for the principle of proportionality. § 32f(4) of the Draft Bill allowed for the remedy of divestments only as a last resort and gave the Monopolies Commission and the competition authority in the Land where the undertaking had its seat the opportunity to comment on the Bundeskartellamt’s draft order before it was issued. Where the assets to be divested had been acquired based on a final decision of the Bundeskartellamt or the European Commission or a ministerial exemption, divestiture could not be ordered before five years had lapsed since the decision (or, in the absence of Phase II-referral in Germany, the lapse of the time period for referrals).[5]

In lieu of a remedies order, the Bundeskartellamt could make commitments offered by the undertaking binding on them (§ 32f(6) of the Draft Bill).

In addition to the introduction of the ‘German NCT’, the Draft Bill made a few minor changes in the context of sector enquiries. First, the trigger for sector enquiries is currently that ‘rigid prices or other circumstances’ raise the suspicion that competition could be restricted or distorted. The Draft Bill deletes the words ‘rigid prices or’. Second, the Draft Bill introduces obligations that the sector enquiry should generally be concluded within 18 months, and that any subsequent remedies orders under § 32f(3), divestment orders under § 32f(4), or orders to notify future concentrations under § 32f(2) should be issued within 18 months after publication of the final sector enquiry report. However, the 18-month deadlines are not strictly binding, and there are no consequences attached to non-compliance with these deadlines. Third, the Draft Bill empowers the Monopolies Commission to make recommendations for starting a sector enquiry. While this recommendation is non-binding for the Bundeskartellamt, the Draft Bill requires the Bundeskartellamt to explain its reasons for not starting a sector enquiry if it has not done so within six[6] months of the publication of the recommendation.

2. Disgorgement of benefits

An additional, but more incremental, aim of the Draft Bill was to render more effective the remedy of disgorging the economic benefits of an undertaking that had infringed competition law.

Ein Vogel fliegt am Himmel
Bundeskartellamt will take those profits away from you!

For more than four decades, the ARC has empowered German competition authorities to disgorge economic benefits from undertakings that have intentionally or negligently infringed competition law. The provision was originally introduced in 1980, then as § 37b ARC 1980, and renumbered as § 34 in 1998, where it is still found today. The provision has achieved hardly any practical significance.

Alternatively to § 34, the competition authority may disgorge the benefits by increasing the fine by the amount of the economic benefit where the competition authority issues a fining decision.[7] Competition authorities, however, avoid splitting the fine into a punishing and a disgorging part because it would require them to quantify the economic benefits caused by the infringement. There is a suspicion that competition authorities instead implicitly calculate the fine such that economic benefits of the infringement are disgorged, even when they profess the fine to contain purely punishing elements.[8]

The background to the renewed interest in this provision on disgorgement of benefits is that the Russian war of aggression against the Ukraine and the energy crisis led to a sharp rise of oil and petrol prices, and to a sharp rise of profits of some undertakings in the energy sector. As always when petrol prices rise, the public and politicians call for competition law interventions – a call that ultimately tends to remain unheard because it is usually misplaced where a shock to demand or – as here – a shock to supply leads to higher prices. Be that as it may, ‘disgorging benefits’ sounded just like the instrument to help, and so the attention turned to § 34 ARC. However, the provision did not apply to this scenario for several reasons; and – spoiler alert! – even after the now proposed changes, it still would not have applied to the scenario of rising prices in the energy sector to the extent they were caused by a shock to supply without an underlying infringement of competition law.

The Draft Bill sought to make § 34 ARC more effective by proposing three changes: first, deleting the reference to ‘intentional or negligent’ infringements;[9] secondly, adding a presumption that the infringement of competition law leads to economic benefits; and thirdly, adding a second presumption that such an economic benefit amounts to at least one per cent of the relevant turnover.

3. Public and private enforcement of the DMA

The third set of changes concerned the public and private enforcement of the DMA.

  1. Public enforcement of the provisions of the DMA is concentrated with the European Commission. However, the DMA leaves some space for support by national competition authorities. The Draft makes use of this space by introducing a new § 32g, extending the Bundeskartellamt’s investigative powers to cases where non-compliance with Articles 5, 6, or 7 of the DMA is suspected, and by extending the jurisdiction of the Bundeskartellamt (§ 50) and the rules on cooperation with other authorities (§ 50f ARC) accordingly.
  2. In addition, the Draft Bill creates a private cause of action for damages where Articles 5, 6, or 7 of the DMA are infringed, and extends the procedural provisions on private competition law enforcement (including those on the binding effect of infringement decisions, limitation periods, and the substantive and procedural rules on disclosure) to actions for non-compliance with Articles 5, 6, or 7 of the DMA.

II. Criticism against the Draft Bill of September 2022

The critics against the Draft Bill fell into two camps: those who fundamentally opposed the proposals (especially the German NCT) and those who supported the German NCT in principle but took issue with the details.

1. Fundamental opposition to the proposals

The consultation process revealed general hostility towards the proposals in the Draft Bill on the part of many industry associations. The harshest criticism is directed at the German NCT and the possibility of divestments in § 32f of the Draft Bill. The idea that any undertaking could come into the crosshairs of the Bundeskartellamt without infringing competition law and that the Bundeskartellamt could impose any remedy, including the break-up of undertakings, in case of an ill-defined malfunctioning of competition is labelled as a paradigm shift, a move to an arbitrary market design by the state, and – in the more extreme comments – a move to a planned economy. Some members of the academic community voice similarly fundamental, though generally not quite as drastically worded, criticism. At a minimum, the introduction of a market investigation reference regime moves away from the current paradigm of only addressing restraints of competition, and many commentators consider this to be an objectionable fundamental change of paradigm.

Mann im schwarzen Mantel steht tagsüber in der Nähe des Black Metal Gate
Police at 10 Downing St (headquarters of planned economy to some).

2. Qualified support with constructive criticism

Other critics were less fundamentally opposed to the proposal or welcomed the proposals with open arms. Yet even those who support the introduction of the German NCT in principle (full disclosure: Yours truly included) criticized the Draft Bill for its lack of guidance it gave to the addressees on what was to be understood under the new concept of a ‘malfunctioning of competition’. The criticism conceded that a crisp and comprehensive definition of a general clause that is meant to act as a gap filler is not possible and that the exact contours of the concept will necessarily remain blurry. However, this should not have prevented mentioning case categories or sample theories of harm. In the UK, where the ‘definition’ of the ‘adverse effect on competition’ (AEC) is similarly elusive as the German Draft Bill’s ‘malfunctioning of competition’, the Competition Commission’s guidance (adopted by the CMA) points out structural and conduct features, and identifies as potential sources of competitive harm: 1. unilateral market power; 2. barriers to entry and expansion; 3. coordinated conduct by firms, 4. vertical relationships, and 5. weak customer response.[10]

Well-meaning critics of the German Draft Bill suggested that a similar catalogue of theories of harm and a corresponding non-exhaustive list of examples would go a long way towards attenuating the industry’s concern about legal certainty. Further suggestions for improving the Draft Bill’s approach based on the UK experience included a call for clarification that the NCT should only be used where enforcement of the conduct rules was not considered sufficiently promising, and a two-stage, ‘fresh-pair-of-eyes’ investigation and decision-making process similar to the UK regime (either of the institutional variety, as was the case before the ERRA2013, or of the internal variety, as has been the case within the CMA since the ERRA2013). If these measures should be insufficient to alleviate the industry’s concerns, it was proposed that safe harbours could be considered, eg, to exclude markets where a malfunctioning of competition is unlikely or where the effects would only be de minimis.

III. The Government Bill of 5 April 2023

As mentioned in the introduction, the Government Bill was published and presented on 5 April 2023. The Government Bill is in large part and in its main thrust identical to the Draft Bill but puts some flesh on the bare bones of the concept of the ‘malfunctioning of competition’. It also introduces a number of procedural safeguards and makes some minor changes to the Draft Bill and its Explanatory Note.

klares Glaskuppelgebäude
The German parliament (the Bundestag) now has to deal with the new competition rules.

1. Changes to the German NCT in § 32f

  • The Draft Bill’s wording of the central condition for intervention, the ‘substantial, continuing or repeated malfunctioning’ (erhebliche, andauernde oder wiederholte Störung), had been criticised: Even if the intended meaning was relatively clear, it seemed to put the three adjectives in an alternative relationship. For example, under a literal interpretation of the Draft Bill’s wording, an insubstantial but repeated malfunction would have been caught. The Government Bill now substitutes the unambiguous formulation ‘substantial and persisting malfunction’ (erhebliche und fortwährende Störung) for the Draft Bill’s ambiguous wording.
  • At the same time, the Government Bill now clarifies in a legal definition that a malfunctioning is ‘persisting’ if it has existed for at least the past three years and there are no indications that it will disappear, on the balance of probability, within the next two years (§ 32f(5)3 in the version of the Government Bill).
  • There is no definition for ‘substantial’ in the operative part of the Government Bill. The Explanatory Note to the Government Bill (at p. 31) explains that a malfunctioning is ‘substantial’ if it ‘has more than merely de minimis effects’. This is a surprisingly low bar for intervention, and not necessarily a bar one would associate with ‘substantial’. It is true that in some contexts, for example when it comes to the qualified effects test in international competition law, ‘substantial’ is used for what is essentially a mere de minimis threshold. And yet, even though I am fairly trigger-happy when it comes to the enforcement of competition law, in this context, where intervention is not based on an infringement of a discrete prohibition but only on the objective malfunctioning of competition, I would have expected a higher bar for intervention, and ‘substantial’ usually does denote a higher bar than a de minimis standard.
  • The Government Bill introduces the qualification that the substantial and persisting malfunctioning must affect competition on one at least nation-wide market, several markets, or across markets. Given that these are alternatives, and the size of the market volume or the size of several sub-national markets remains undefined, this does not seem to be a high hurdle to overcome.
  • Paying heed to the criticism mentioned above (II.), the Government Bill now includes not only the (slightly modified) list of factors to be considered in finding a malfunctioning, but also a non-exhaustive list of examples of categories of theories of harm where a malfunctioning is ‘in particular’ to be found. Similar to the list in the Competition Commission’s guidance (CC3 revised), these categories of theories of harm are: 1. unilateral market power in supply or demand; 2. barriers to entry, barriers to exit, restrictions of capacities or barriers for consumers to switch suppliers; 3. parallel or coordinated conduct; 4. input or customer foreclosure through vertical relationships.[11] The additional list of factors to be considered in finding a malfunctioning now also includes transparency and homogeneity of the goods on the relevant markets, but also ‘mitigating’ factors such as the degree of innovation on the markets and efficiencies (provided consumers receive an appropriate share).[12]
  • The Government Bill now includes a condition that, on the basis of the facts as understood at the time of decision-making, enforcement of the conduct prohibitions against anticompetitive agreements (§ 1 ARC) or abuses of market power (§§ 18 et seq. ARC) must likely be insufficient to counter the malfunctioning appropriately, thus introducing a ‘soft subsidiarity’ of the application of the German NCT vis-à-vis the existing conduct prohibitions.
  • The Government Bill now excludes remedies orders on the basis of sector enquiries based on repeated infringements of consumer law.[13] It is not entirely clear what this would mean where such a sector enquiry finds that these infringements result in, for example, a barrier for consumers to switch suppliers, and therefore a malfunctioning of competition.
  • With regard to procedure, the Government Bill has made major changes. The process is now divided into three steps.
    • First, there is the sector enquiry, which must precede any finding of a malfunctioning.
    • Second, if the Bundeskartellamt intends to make a finding that there is a substantial and persisting malfunctioning of competition, it is required to issue a corresponding decision and to address this decision to one or more undertakings which, by their conduct, substantially contribute to the malfunctioning of competition.
    • Third, the Bundeskartellamt may, following an oral hearing in which the parties and the Monopolies Commission have a right to be heard,[14] order remedies or make commitments made by the undertakings binding on them.

The second step, newly introduced in the Government Bill, could delay the process by more than it may at first appear. It allows the addressees to appeal the decision containing the finding of a malfunctioning. While such an appeal does not have suspensive effect, the President of the Bundeskartellamt Andreas Mundt has already indicated[15] that it seems very unlikely that the Bundeskartellamt would proceed to the third step before the decision about the finding of a malfunctioning has become final.

  • The non-exhaustive list of possible remedies now contains the prohibition of unilateral information disclosures where such disclosure could facilitate collusion, which had been conspicuously absent in the Draft Bill.
  • The provision on divestment orders (§ 32f(4)) has also been significantly reformed. First, where the Draft Bill allowed divestment orders to be issued against any undertaking (provided this contributed to the elimination or substantial improvement of the malfunction), the Government Bill restricts such orders to undertakings in a dominant position or undertakings of paramount significance for competition across markets. Second, procedurally an oral hearing in which the Monopolies Commission has a right to be heard is required before the order. Third, the Government Bill now assigns suspensive effect to appeals against the divestment order. Fourth, the shares or assets only need to be divested if the price offered by the purchaser amounts to at least 50 per cent of the value as assessed by an accountant appointed by the Bundeskartellamt. Fifth, the Government Bill now takes up a suggestion made by the Monopolies Commission and provides for state compensation in some circumstances: where the purchase price offered is lower than the value as assessed by the accountant (but 50 per cent or more than this value), the state will pay 50 per cent of the difference between the value and the purchase price. This means that the addressee of the divestment order will at a minimum receive 75% of the assessed value of the shares or assets.
  • Where the shares or assets to be divested had been acquired by way of a concentration reviewed by a competition authority and the concentration had been cleared in a final decision by the European or German institutions (or where the Bundeskartellamt decided not to enter Phase II, or where there was a ministerial exemption for the concentration), a divestment order cannot be issued for at least 10 years since the date of the decision (or, mutatis mutandis, the date in which the period for the decision to enter Phase II expired).
  • The Government Bill now provides in § 32f(8) that in the case of regulated sectors (railways, postal services, telecommunications) or regulated electricity or gas networks, the Bundeskartellamt may order remedies under § 32f(3) or order divestments under § 32f(4) only where the Federal Network Authority (Bundesnetzagentur) gives its assent. Implicitly, this clarifies that the rules are applicable to these regulated sectors and networks as well.
Next stop: Divestiture?

2. Changes to the disgorgement of benefits

Compared to the significant changes to the NCT in § 32f, the changes to the Draft Bill’s rules on disgorging benefits are minor. The Government Bill no longer proposes to delete the requirement that disgorgement is only possible for ‘intentional or negligent’ infringements. Considering that the disgorgement of benefit is not a rule on damages, but more akin to unjustified enrichment, where we do not require fault for liability, one could quibble with this change. However, given the very low threshold for negligence where competition law infringements are concerned, this is not likely to make any practical difference. The only other change of the Government Bill’s wording compared to the Draft Bill is the introduction of a legal definition for the ‘period of disgorgement’ of five years in § 34(5), and the corresponding rephrasing in § 34(4).

3. DMA enforcement

The only (minor) changes between the Draft Bill and the Government Bill with regard to the changes to the public and private enforcement are the amendments

  • that where an infringement of Article 7 DMA is considered possible, the Bundeskartellamt will give the Federal Network Authority an opportunity for comment;
  • that the binding effect on courts is extended to the European Commission’s designations under Article 3 DMA; and
  • of the full citation to the title of the DMA.

IV. My own two cents: Comments on the Government Bill

From the outset, I considered the proposals in the Draft Bill, in particular the introduction of the NCT in § 32f, to be welcome additions in the armoury of the Bundeskartellamt.

1. Fundamental criticism against the NCT unjustified…

Some of the commentary on the German NCT clearly is over the top. Invoking a ‘planned economy’, when the NCT does no more than essentially copy the UK’s regime of market investigation references in Part IV of the Enterprise Act 2002, is hyperbole far removed from serious discourse. There are quite a few things I would call David Cameron, Theresa May, Rishi Sunak or, in particular, Boris Johnson and Liz Truss with her ‘Britannia Unchanined’ co-author Kwasi Kwarteng as Chancellor, but ‘socialist’ is not necessarily the word that comes spontaneously to mind.

I also did not, and still do not, share the fundamental criticism that the NCT is incompatible with the ordoliberal approach. The original ordoliberal approach, as proposed in the Josten Draft of 1949 (co-authored by Franz Böhm), would have eliminated any economic market power not achieved by pioneering initiatives. It would have done so by breaking up entities with market power if the ‘position of power could not be otherwise eliminated on a permanent basis’ (§ 15 of the Josten Draft) – in other words, the Josten Draft did include the possibility of divestments (or other remedies) even in the absence of infringing conduct. This approach was suggested not because the drafters did not know any better: they deliberately chose to address the issue of market power at its roots rather than address market power only once it was abused.[16] Market interventions would have been even more radical under Walter Eucken’s proposals for as-if-competition.[17]

Following intense lobbying by the German Industry Association BDI, the Josten Draft never became law. Instead, under the ARC as eventually passed, existing positions of market power were left unaffected, and the decision not to include merger control in the original ARC led to further unchecked concentration. Starting from a skewed playing field, simply invoking von Hayek’s competition as a discovery process does not guarantee that efficient competitors face contestable markets. Intervening where markets are not contestable in order to open up markets to competition, even in the absence of restrictive agreements or discrete abusive conduct, hardly goes against the teachings of Eucken and Böhm.

It is true that the introduction of the German NCT means that we are moving from a system characterized by deficient coverage – leaving gaps such as tacit collusion with or without facilitating practices – to a system including a general clause that could potentially be stretched to cover too much or to allow remedies that go beyond that which would be necessary. Reasonable minds can differ on which side to err. However, there have always been regrettable gaps in the enforcement armour, in particular when it comes to oligopolies. Newer developments mean that these gaps are widening: Algorithmic collusion and the parallel use of the same algorithm will exacerbate the problem of tacit collusion. Digital conglomerates with their ecosystems foreclose large parts of the economy to dynamic newcomers because of network effects and advantageous access of data. Developments in the financial world mean that new gap cases may come into existence, such as potential problems with common ownership, or old ones may become more frequent, such as minority shareholdings in competitors or interlocking directorates. Given these developments, I consider it preferable to err on the side of a competition law instrument that allows for intervention.

Sven Giegold, former Member of the European Parliament turned State Secretary, led the drafting process of the 11th amendment to the German competition act.

2. … but calls for more detail than in the Draft Bill were justified – and have been heard

That is not to say that one need not be careful when intervening on the basis of an instrument like the German NCT. Where there are competition law infringements, the trigger for intervention is (moderately) well defined, and remedies can be surgically tailored to removing the infringement. In contrast, interventions without infringements, based on a general malfunctioning of competition, do not put similarly defined boundaries on remedies. This must not result in the competition authority designing markets by doing more than removing the obstacle identified. This danger needs to be borne in mind, but this is also true under the already existing law in the many cases in which commitment decisions are issued.

The German NCT as it was designed in the original Draft Bill of September 2022 gave very little contextual information, and instead provided for a general clause with ill-defined contours. Based on the Draft Bill, it would have been difficult for undertakings to foresee in which circumstances they might face interventions, and difficult for courts to assess whether remedies imposed were substantively legal because there was no yardstick against which to measure legality; appeals would arguably only have been restricted to a proportionality review. The Draft Bill also lacked procedural safeguards. In essence, one had to trust the Bundeskartellamt to practice self-restraint. While such trust would arguably be justified,[18] it has to be conceded that it is best not to build on trust alone when an institutional set-up is designed.

The Government Bill has paid heed to the criticism levied against the Draft Bill and taken great pains to address the deficiencies of the Draft Bill (above III.). Of course, not all uncertainty has been removed from the German NCT. Some uncertainty is inevitable where a general clause is meant to fill the enforcement gaps in a dynamic environment. However, § 32f(5) with its list of examples and factors to be considered, while non-exhaustive, gives a clear idea what circumstances may result in interventions based on what theories of harm. The close modelling of the German NCT on the British Part IV regime allows drawing parallels to the corresponding case law for orientation. The level of detail of § 32f in the Government Bill’s version is sufficient for a court to assess whether the Bundeskartellamt has made a credible case that a malfunctioning exists, and to assess whether the remedies are proportionate to the aim of eliminating that malfunction. The restriction that the Bundeskartellamt may choose as addressees only those undertakings that contribute substantially is a sensible focusing of enforcement energy, and at the same time removes the anxiety of smaller players that they could be caught in the net.

3. Yet more room for improvement? The case for a fresh pair of eyes

There is one additional procedural safeguard I would have liked to see that is not (yet?) to be found in the legislative proposal: the requirement for a ‘fresh-pair-of-eyes’ approach such as the one in the institutional set-up in the UK before the ERRA2013 or the one that is now replicated within the CMA. The necessarily vaguer concept of a malfunctioning of competition compared to a competition law infringement leaves greater leeway for differences of opinion whether a competitive situation amounts to a malfunctioning and whether intervention is justified (and if so, what the appropriate remedies are). There is a danger of commitment bias and confirmation bias on part of even a well-meaning enforcer where that enforcer has spent substantial time and effort in investigating a market or markets.

Ein Lemur sitzt auf einem Geländer
Let’s look at this with a fresh pair of eyes.

It is true that there is the possibility of appealing the Bundeskartellamt’s decision to the courts, so that there is a ‘fresh pair of eyes’ at that stage. The Government Bill has strengthened this possibility by allowing appeals against the finding of a malfunctioning at the second stage even before remedies are ordered at the third stage. Yet, the softer boundaries of the concept of a malfunctioning mean that the courts will arguably pay greater deference to the competition authority’s institutional advantages. Courts are institutionally not well-suited to replicate the broad investigation into an entire sector. While there is no formal limitation to the judicial-review standard (as is the case in the UK), de facto the yardstick under the German regime will not be very different. At the same time, the softer boundaries of the concept of malfunctioning of competition also mean that court supervision of remedies will not be as strict as in the case of infringements of competition law (although supervision will still be stricter than in the case of commitment decisions, because there the argument is that the parties agreed to the remedies[19]).

For these reasons, I would have preferred a fresh-pair-of-eyes approach before the administrative decision is issued. The Josten Draft of 5 July 1949 provides an interesting three-stage blueprint. It suggested that before a divestment order, at the first stage the undertaking should be given three months to propose a plan. If this plan was not forthcoming, or if it was not satisfactory, the competition authority should then propose its own plan to the Minister for Economic Affairs. If the Minister agreed with the plan, it would be implemented. If the Minister did not agree, the competition authority’s plan would be put to a select committee of the German Bundestag; if it agreed, the plan would be implemented.

The first stage of the Josten Draft is already part of § 32f, because the rules on commitment decisions (§ 32b ARC) are included by reference. It seems very likely that in practice many, if not most, § 32f investigations will end in commitment decisions, if only because the breadth of potential remedies in a remedies order gives the competition authority a powerful threat point.

The introduction of the second and third stages of the Josten Draft could, first, address the problem of commitment and confirmation bias mentioned above; they would provide a ‘fresh pair of eyes’. This was the reason why the Josten Draft proposed them. Additionally, the requirement of consent of political actors (either the Minister or the select committee) would acknowledge the fact that the new instrument is one that straddles the line between competition law enforcement and regulation.

The proposal faces three potential objections: procedural ineffiency; reduced incentives for the competition authority to expend resources on investigations; and politicization of competition law.

First, any ‘fresh-pair-of-eyes’ approach necessarily brings with it costs in the form of delays, effort, and money. This is why in the UK the institutional duplication of the investigations by the OFT and the Competition Commission was abandoned in the ERRA2013, and why the intra-institutional duplication within the CMA was softened by allowing individuals from the Phase I team into the Phase II team. Ideally, the second stage should not try to replicate the entire decision-making process but should restrict itself to a plausibility and appropriateness analysis with tight time-limits. The additional delay should not be such as to deprive the NCT of its practical effectiveness.

Second, a review of the decision by another actor may reduce the Bundeskartellamt’s willingness to expend substantial resources on an investigation if there is a danger that these resources will be wasted where assent by the second-stage decision maker is not forthcoming. However, the Josten Draft’s proposal alleviates this concern by giving the competition authority three bites at the apple by requiring alternatively the assent by the undertaking, the Minister, or the select committee.

The third potential objection is not a general one against a fresh-pair-of-eyes approach but specifically one against participation of the Minister or the select committee in the decision-making process. Such participation of political actors may lead to a politicization of competition law. The German experience with ministerial exemptions in merger cases has indeed not been a happy one. The problem is arguably less problematic when it comes to the German NCT, because this instrument straddles the lines between competition law enforcement and regulation anyway. The Bundeskartellamt would retain the power of initiative and would be the only player able to design the content of the remedies order. In other words, it would remain primus inter pares among the decision makers; the Minister or the select committee should only be able to give assent to the draft order (vel non), but not to influence its content.

V. Likelihood of the Government Bill passing into law

Despite the not insubstantial blowback against the Draft Bill and to a lesser extent the Government Bill, it seems likely that the NCT proposal will pass into law. Two of the three coalition partners have nailed their flags to the mast. The initial bill was drafted by the Ministry for the Economy and Climate Protection under leadership of a Minister from the Greens. It seems unlikely that the Greens in Parliament will be impressed by the criticism against the bill. For those on the left wing of the Greens, state intervention is not a problem to begin with, and for those on the moderate wing of the Greens, the intervention to strengthen competition in non-functioning markets is a way to raise consumer welfare while respecting the logic of a competitive market economy.

The liberal party (FDP) has, in principle, also been a supporter of the proposal. They had, while in a different coalition with the Christian Democrats and under a different Minister for the Economy (Rainer Brüderle, FDP), introduced a proposal for the introduction of a power to order divestitures even in the absence of a competition law infringement back in the year 2010. This proposal had failed, among other things because their then-coalition partner CDU/CSU did not sufficiently support the proposal and because the proposal did not provide for any milder means; given that divestitures are a last resort only and very unlikely to be ordered, the 2010 bill had been considered largely a symbolic one.

This background is likely to strengthen the FDP’s resolve to make the current Government Bill succeed in Parliament. Both Robert Habeck (Greens) and Marco Buschmann (FDP) emphasized the ‘origin story’ of the new competition tool being the direct descendant of the Brüderle Draft of 2010 in the press conference presenting the Government Bill on 5 April 2023. The FDP in Government has therefore committed to the bill, not by grudging support for a proposal by the Greens, but by taking (co-) ownership. Invoking the Brüderle-Draft also means that a renewed failure to pass the Bill would result in a political embarrassment to the FDP. Nevertheless, it is possible that the FDP in Parliament may be in two minds about the Government Bill. The FDP has always been torn between a liberal instinct to protect free and undistorted competition on the one hand and a certain proximity to industry associations on the other. It seems likely that FDP support for the Government Bill in Parliament will not be undivided, and it is quite possible that the FDP will seek to make amendments to the Government Bill (possibly by introducing a fresh-pair-of-eyes approach?).

The SPD has not made any similar public precommitment to the Bill, except for not stopping the Government Bill from passing Cabinet scrutiny. The coalition agreement between the SPD, Greens, and FDP had provided for support for a power to order divestitures independently of a finding of an infringement of competition law, but only on the European level.[20] While the SPD is therefore not equally invested in the project as the Greens or the FDP, it seems likely that they will support the project.

Of course, it is difficult to predict whether modifications will be made along the legislative process, and if so, of what kind. It is to be hoped that the German NCT will become law and that any amendments to the bill will not deprive the instrument of its practical effectiveness.

Bild Wagner-von Papp

Prof. Dr. Florian Wagner-von Papp is Professor of Private and Business Law at the Helmut Schmidt University/University of the Federal Armed Forces in Hamburg (HSU) with a research focus on antitrust/competition law, comparative law and contract design. He joined HSU in 2019 from University College London (UCL) where he had been Professor for Antitrust and Comparative Law & Economics (and where he had been teaching since 2005). 


[1] I have used the phrase ‘malfunctioning of competition’ here because this is how officials from the Ministry for the Economy and Climate Protection translated the German phrase ‘Störung des Wettbewerbs’. The phrase sounds a little clunky. I would be tempted to use ‘distortion of competition’ as a more elegant translation for ‘Störung des Wettbewerbs’; but this sounds as if the new provision used the formulation contained in Article 101(1) TFEU, while the German phrase in Article 101(1) is a different one, namely ‘Verzerrung des Wettbewerbs’.

[2] This part of the provision (Draft-§ 32f(2) ARC) is a slight modification of the current § 39a ARC, which had already been introduced in the last Amendment to the ARC.

[3] The draft explanatory note gives full credit to both these role models.

[4] United States v. Aluminum Co. of America, 148 F.2d 416, 430 (2d Cir. 1945).

[5] In the Government Bill, this has been extended to 10 years since the decision had become final.

[6] In the Government Bill, this has been extended to 12 months.

[7] §§ 81 et seq. ARC provides for administrative fines for infringements, and § 81d(3) ARC provides that § 17(4) of the Act on Administrative Offences applies. The latter provision makes disgorging the benefits of an offence mandatory. However, § 81d(3) ARC modifies this and gives the competition authority the choice either to include a disgorging element to the fine (in which case disgorgement under § 34 ARC can no longer be ordered to the extent the fine disgorges the benefits, § 34(2)1 no. 2 ARC), or to impose a fine purely for purposes of punishment (In which case the economic benefit may be disgorged by way of § 34 ARC).

[8] Wouter P.J. Wils has shown that the European Fining Guidelines are an approximation for disgorging the economic benefits. Wils, Wouter P. J., The European Commission’s 2006 Guidelines on Antitrust Fines: A Legal and Economic Analysis. World Competition: Law and Economics Review, Vol. 30, No. 2, June 2007, Available at SSRN: While the German Fining Guidelines differ from the European ones, the logic would appear to be similar. Kersting, in: Kirk, Offergeld &Rohner (eds), Kartellrecht in der Zeitenwende, Nomos 2023, 79, 80f., argues that this implicit disgorgement would be illegal, inter alia, because it would deprive the infringer of the possibility to deduct the disgorging part of the fine for tax purposes.

[9] This proposal is no longer included in the Government Bill.

[10] Competition Commission, Guidelines for market investigations: Their role, procedures, assessment and remedies (April 2013) CC3 (revised). These were adopted by the CMA.

[11] § 32f(5)1 of the Government Bill.

[12] § 32f(5)2 of the Government Bill.

[13] § 32f(1)2 in the Government Bill’s verssion.

[14] § 56(7) in the Government Bill’s version

[15] Presentation on 25 April 2023 at the 4th Competition Day of the Markenverband e.V.

[16] Walter Bauer, Franz Böhm, Durt Fischer, Paul Josten, Wilhelm Köppel, Wilhelm Kromphard & Berhard Pfister, Entwurf zu einem Gesetz zur Sicherung des Leistungswettbewerbs und zu einerm Gesetz über das Monopolamt mit Stellungnahme des Sachverständigen-Ausschusses und Minderheitsgutachten, presented to the Director of the Administration of the Economy Ludwig Erhard on 5 July 1949, pp. 119-120.

[17] Eg, Walter Eucken, ORDO 2 (1949) 68.

[18] Critics hastened to add that they trusted Andreas Mundt, but often added that one did not know who would follow him as the next President of the Bundeskartellamt. This reveals a misunderstanding of the President’s role within the Bundeskartellamt. The Bundeskartellamt’s decisions are handed down by independent Decision Panels, without influence by the President.

[19] Whether that argument is persuasive is a separate question. For my views see Florian Wagner-von Papp, ‘Best and Even Better Practices in Commitment Decisions after Alrosa: The Dangers of Abandoning the Struggle for Competition Law’ (2012) 49 CMLRev 929.

[20] Additionally, the coalition agreement had envisioned empowering the Bundeskartellamt to investigate and impose remedies where ‘substantial, persistent or repeated infringements of consumer protection law’ occurred (a formulation that closely resembles the wording of the Draft Bill of August 2022, where it was the ‘substantial, persistent or repeated distortion of competition’ that allowed intervention). Curiously, the Government Bill, which allows the imposition of remedies where a ‘substantial continuing or repeated distortion of competition’ is found, excludes sector enquiries that are based on repeated infringements of consumer protection laws.

2 thoughts on “The 11th Amendment to the ARC and Germany’s New Competition Tool

Leave a Reply

Your email address will not be published. Required fields are marked *