One of the issues which has bedevilled agreement at EU level on mechanisms which might allow a company to alter its applicable corporate law (without changing anything else) is the potential impact of this choice on employee participation rights at board level. Although not crucially raised at the point of incorporation (ie in the classic Centros situation) because these companies typically operate below the participation thresholds, the issue is fully engaged with mid-stream re-incorporations. The background problem is that Member States’ rules on board level participation are diverse, both among those states which have them and in the sense that about half the Member States do not have general board level participation rules at all. And all Member States have shown themselves very resistant to harmonisation of the rules in this area. Harmonisation of board structures and composition was something which the Commission attempted in the draft Fifth Company Law Directive, proposed by the Commission in 1972 and eventually abandoned for lack of agreement among the Member States nearly thirty years later (in 2001). Whereas sub-board participation requirements, whether by way of works councils or collective bargaining, are regarded as labour law and apply on a territorial basis (ie to all operational activities in a particular jurisdiction, whether the company is incorporated in that jurisdiction or not), board level participation rules are embedded in corporate law and so at risk of being circumvented by a change in that law.

The solutions actually adopted in the EU essentially qualify the free choice of applicable company law. One approach is to make mandatory the ‘real seat’ test for valid incorporation. Thus, a European Company (SE) must be incorporated in the jurisdiction in which it has its head office (Art 7 of the SE Statute). Here the company adopts the whole of the company law of choice, but at the cost of shifting its headquarters to that jurisdiction, a cost it may not wish to incur. However, this was not a solution which could be used in the Cross-border Mergers Directive of 2005 or the recently adopted Cross-border Conversions Directive, at least it was not one which incorporation theory states were prepared to accept since it entailed a change in their domestic rules on valid incorporation. This comment focusses on these two Directives. The 2019 Directive on Conversions also amended the earlier Cross-Border Mergers Directive. The Mergers Directive in the meantime had been consolidated in Title II of Directive (EU) 2017/1132 (hereafter the Consolidating Directive), into which the new Conversions rules have now also been inserted. The text of the 2019 Directive referred to is that which appears on the website of the European Parliament.

Both Directives open up the possibility of a transfer of jurisdiction driven by pure choice of law considerations. In the case of a merger, it is necessary for this purpose to set up a shell company in the jurisdiction of choice and merge the operating company into it. However, the merger may also involve the integration of a number of operating companies and be driven by the goal of eliminating unnecessary corporate vehicles as well as changing the operative corporate law. Or change of corporate law may not be a driver of the transaction, but the merging companies still have to settle on a single jurisdiction for the resulting company. A Conversion by contrast involves only a change of operative corporate law: it is ‘an operation whereby a company, without being dissolved, wound up or going into liquidation, converts the legal form under which it is registered in a departure Member State into a legal form of the destination Member State […] and transfers at least its registered office into the destination Member State whilst retaining its legal personality.’ (Art 86b(2) of the Consolidating Directive)

For the Directives, the second solution was the important one, ie to accept diversity at the cost of inhibiting the company from opting for the whole of the company law of its jurisdiction of choice. This solution was worked out for the European Company statute and then rolled out for the Mergers and Conversions Directives as well (so that the SE Directive 2001/86/EC provides the basic rules for mergers and conversions as well). It is sometimes referred to as the ‘no escape, no extension’ rule. In principle, the company is bound by the employee participation rules (if any) of the transferee jurisdiction, but if the company is already subject to participation rules in the transferor jurisdiction, it will become subject to a similar set of rules, now incorporated into the transferee jurisdiction (at least if the transferor jurisdiction rules are ‘stronger’ than those operating in the transferee jurisdiction). So, if two or more companies merge across borders or if a company converts to another jurisdiction and none of them is subject to participation rules in the transferor jurisdiction, then those companies become subject to participation rules only if the transferee jurisdiction has them (and, if so, the companies can be said to have freely chosen them). This is ‘no extension’.

The ‘no escape’ concept was more difficult to implement. In principle, if the transferring company was subject to participation rules before the transfer the same rules should apply afterwards, but this was not possible to achieve in all cases. Sometimes escape possibilities had to be allowed; sometimes extension ones. Four big problems arose. First, there is the problem of comparing apples and oranges, given the variety of participation schemes operating in those Member States which have them. ‘Participation’ is defined as the ‘influence’ employees have on the board, whether by way of electing or appointing board members or the right to recommend or oppose the appointment of board members. (SE Directive, art 2). These mechanisms are treated equally and the ‘strongest’ mechanism is the one by which the employees exercise influence over the highest proportion of board members (SE Directive, Annex, Part 3). So, the right to make recommendation over half the board trumps a right to appoint one third of the board members – an odd functional result.

Second, there is the limited geographical scope of national participation rules. They tend to embrace only the employees of companies incorporated in the jurisdiction and sometimes only if those employees work in the jurisdiction as well, ie not employees of subsidiaries and establishments located in other member states. Not at all communautaire. Consequently, what is applied in the appropriate cases is not the rules of the transferor jurisdiction as such, but ‘standard rules’ created by EU legislation and modelled on the ‘strongest’ participation model, but extended to the EU employees of the whole group headed by the transferring company. (SE Directive, Annex, Part 1). The same rule is applied to the transferee jurisdiction’s rules if these are the applicable ones and are jurisdictionally confined. This solution puts the settlement of disputes into the hands of the relevant bodies in the transferee state. Those bodies may or may not do a good job of applying participation rules with which they are unfamiliar.

Third, there is the problem of the tail wagging the dog. In the case of a cross-border merger, there may be multiple companies involved, with different participation systems. The outcome of applying the ‘strongest version’ test may be that the resulting company is subject to an employee participation system which previously applied only to a small proportion of the total workforce of the resulting company. The Directive offers a number of ways of dealing with this issue, but they all depend upon negotiations with representatives of the employees who will work for the resulting company and its subsidiaries (who are formed into a ‘special negotiating body’ (SNB) for this purpose).

(i) If the percentage of the post-merger workforce who are covered pre-merger by any form of participation system is below 33%, the standard rules do not apply unless the SNB decides to apply them. (SE Directive, art 7(2)(b); Consolidating Directive art 133(3)(e)) This appears to be a decision it can take by an absolute majority of its members representing an absolute majority of the resulting workforce. This is perhaps most likely where the participation system was confined to a small minority of the resulting workforce.

(ii)The SNB by a two-thirds majority (of both members and employees represented) can opt out of the standard rules altogether or agree with the companies concerned a less demanding form of participation than the strongest one (SE Directive, art 7(4)(6)). A ‘less demanding agreement’ is most likely to be used where the merging companies are subject to a variety of participation schemes, but the strongest one applied previously only to a small proportion of the resulting workforce. Opting out altogether in case not falling within (i) seems unlikely, but there is the possibility that the merging companies could offer something which the SNB would regard as worthwhile for giving up this entitlement.

(iii)Negotiations with the employees, which may last between 6 months and a year, clearly have the capacity to slow down the merger process. However, the merging companies can opt out of the negotiating process, but only at the cost of accepting the standard rules and, thus, the ‘strongest’ version of participation. On the other hand, there is some incentive to go through the negotiating process if the destination member state has taken up the option of limiting the proportion of board seats permitted to be influenced by the employees on a single tier board. The limit is undefined, except that it cannot be less than one third if one of the merging companies was subject to a one third or higher participation scheme. (Consolidating Directive, art 133(4)). The UK limited the proportion to the lesser of one third or the highest previously operative level. Essentially, this means that other Member States do not have to accept the German parity codetermination scheme in the case of cross-border merger – a qualification to ‘no escape’.

(iv) In the case of conversion the multiple prior systems problem does not apply. So, the destination state rules or the departure state rules will apply, whichever are the stronger, but there is still the flexibility to opt out of the standard rules altogether by agreement with the SNB or to negotiate something different from the strongest system. (Consolidating Directive, art 86l(3) and (4)(a))

Finally, there is the big problem of opportunistic use of the transfer mechanisms. Companies might anticipate crossing a national participation threshold (usually defined in terms of the number of employees) and engage in a transfer whilst still participation free (or subject to a lower level of participation). After the transfer, the exiting company is free of the prior participation rules and so can lock itself into what it sees as a favourable status quo. Or the transfer might be a first step in a two-step procedure, under which the second step involves dilution or cessation of the participation rules applied on transfer. The first problem is addressed by bringing within the above rules companies within 80% of the employee threshold on average over the previous six months. (Arts 86l(2) and 133(2) of the Consolidating Directive, as amended in 2019). The second is addressed by requiring the maintenance of the employee participation arrangement for four years (before 2019, three) after the jurisdiction shift, despite any subsequent merger (domestic or cross-border), division or conversion. (Arts 86l(7) and 133(7)). Member states are also required to give the approving authority of the transferring state the right to block mergers and conversions ‘if it is determined in accordance with national law that a [cross-border merger or conversion] is set-up for abusive or fraudulent purposes leading or aimed to lead to evasion or circumvention of national or EU law, or for criminal purposes’ (Consolidating Directive arts 86m(8) and 127(8) – a new provision for mergers). Goodness knows what Member States and their competent authorities will make of this – or what the CJEU will allow them to make - but it could be used in those rare cases where it can be shown that the sole purpose of the transfer is to defeat employee participation rights.

Not surprisingly, no one is happy with this complex set of rules. Those convinced of the value of employee participation regret the even limited possibilities for escape provided by them. See, for example, Martin Gelter, ‘Tilting the Balance between Capital and Labor? The Effects of Regulatory Arbitrage in European Corporate Law on Employees’ (2009) 33 Fordham International Law Journal 792. Those who believe in regulatory competition regret the constraints put on choice of corporate law. It is likely that in practice the rules will produce some division in the market for choice of corporate law. Managers in jurisdictions with employee participation rules seem content with them or at least have worked out to use them to advantage; those in jurisdictions without seem unwilling to become involved with them. This could mean that managers in ‘without’ jurisdictions will prefer to shift, whether by merger or conversion, only to other ‘without’ jurisdictions – and even then only to transferee incorporation-theory states. Managers in ‘with’ jurisdictions will be less constrained. They may not feel inhibited about becoming subject to participation in relation to a larger group of employees (as in a multiple company merger). However, they might be opposed to losing employee participation arrangements they value  as a result of negotiation with the SNB. This might discourage them from becoming minor players in a multiple cross-border merger where the other companies are from ‘without’ jurisdictions.


Paul Davies is a Senior Research Fellow at the Centre for Commercial Law, Harris Manchester College, Oxford.