Friday, September 28, 2012

Yesterday in Luxembourg

AG Bot has another Ruiz Zambrano Case. (I blogged about the three previous ones here.) This time, we have an applicant who lives with his legally resident 3rd national spouse and her EU-citizen child from a previous relationship. AG Bot argues that the applicant can be removed if he does not have sufficient means to support himself. In other words, again the precedent is distinguished in order to keep the rule as narrow as possible. Joined Cases O and S v. Maahanmuuttovirasto and Maahanmuuttovirasto v. L. (NL, DE, FR)

In the category nuisance litigation, we have J. v. Parliament (DE, FR), where the applicant is asking the General Court (Judge Soldevila Fragoso) to annul a decision by the European Parliament declaring his petition under art. 227 TFEU inadmissible on the grounds that it was insufficiently motivated in violation of art. 296 TFEU. This would have been a legitimate complaint, except for the fact that the decision was in fact motivated to the extent that this was possible given the vagueness of the petition. The applicant also tries to get the Court to apply the Charter of Fundamental Rights to the Austrian problem that was the subject of his petition, but no dice, of course. While this is marginally less stupid than asking the Court to review the Parliament’s decision on the merits, I still have to ask: How was this suit not manifestly inadmissible for the purposes of legal aid? (DE, FR)

Otherwise, there were a lot of cartel cases courtesy of Judge Soldevila Fragoso acting as rapporteur. A quick roundup:
  • Flat Glass cartel: Guardian Industries lost and still has to pay € 148 million.
  • Bitumen (NL) cartelShell has a partial win, with the fine being reduced from € 108 million to € 81 million. The Commission failed to prove the alleged aggravating circumstance that Shell was an instigator and leader of the cartel. Nynäs lost and keeps its € 13,5 million fine. Koninklijke Volker Wessels Stevin (NL) and its subsidiary Koninklijke Wegenbouw Stevin (NL, FR) lost and keep their € 27,36 million fine. Kuwait Petroleum lost and keeps its fine at € 16,632 million. Total (FR) and Total Nederland lost and keep their € 20,25 million fine. The Dura Vermeer Group (NL), Dura Vermeer Infra (NL) and Vermeer Infrastructuur (NL) all lost and stay at € 5,4 million. Likewise, Koninklijke BAM lost twice (NL 1, NL 2) and stays at € 13,5 million and Heijmans lost twice (NL 1, NL 2) and stays at € 17,1 million. Ballast Nedam (NL, FR) lost and keeps its € 4,65 million fine. Its subsidiary Ballast Nedam Infra (NL, FR), however, scored a partial win on the grounds that it only became incorporated into the group as of 2000, resulting in a reduction of the share for which it is jointly and severally liable to € 3,45 million. (I.e. of the total for the group of € 4,65 million.)
  • In other words, following 17 cartel judgements authored by Judge Soldevila Fragroso, only the lawyers and Shell achieved a tangible win.


Friday, September 21, 2012

Sans Commentaire

47. Or, eu égard à la situation financière générale extrêmement difficile décrite ci-dessus et à la réaction susmentionnée de la confédération panhellénique des unions de coopératives agricoles, il est hautement prévisible qu’une proportion significative des centaines de milliers de bénéficiaires refuserait de s’acquitter des sommes réclamées, ce qui nécessiterait l’intervention massive des agents de l’administration fiscale, dont le nombre n’a pourtant pas augmenté. Il est évident qu’une telle collecte forcée en masse empêcherait, dans une mesure appréciable, l’administration fiscale de se consacrer à une de ses tâches prioritaires consistant à lutter contre l’évasion fiscale et à collecter des sommes soustraites à l’impôt près de cinquante fois supérieures aux paiements litigieux.

48. S’agissant du risque d’une perturbation de l’ordre public en cas de récupération immédiate des paiements litigieux auprès du secteur agricole grec, il est constant que le climat social en Grèce est actuellement marqué par une détérioration de la confiance à l’égard des pouvoirs publics, par un mécontentement généralisé et par un sentiment d’injustice. En particulier, ainsi que la République hellénique l’a exposé sans être contredite par la Commission, les manifestations violentes contre les mesures d’austérité draconiennes prises par les pouvoirs publics grecs sont en constante augmentation. À l’audition, la République hellénique a encore rappelé la nette progression de certains partis d’extrême droite et d’extrême gauche lors des dernières élections législatives en Grèce.

49. Dans ces conditions, le risque, invoqué par la République hellénique, que la récupération immédiate des paiements litigieux dans le secteur agricole puisse déclencher des manifestations susceptibles de dégénérer en violences n’apparaît ni purement hypothétique ni théorique ou incertain. En effet, il ne saurait être fait fi de la possibilité que l’opération de récupération des paiements litigieux soit publiquement utilisée, par certains milieux, comme exemple de l’injustice exercée contre la classe agricole et que, dans la situation actuelle chargée d’émotions intenses, un tel discours public déclenche l’une ou l’autre manifestation violente, alors qu’il est indifférent de déterminer quelle catégorie de la population pourrait être à l’origine des violences nécessitant un déploiement toujours plus important des forces de l’ordre. Or, il est évident que la perturbation de l’ordre public provoquée par de telles manifestations et par les débordements auxquels les événements dramatiques récents ont montré qu’elles pouvaient donner lieu causerait un préjudice grave et irréparable, que la République hellénique peut légitimement invoquer.

50. Compte tenu des éléments exposés au point 48 ci-dessus, la présente affaire doit être distinguée de celle qui était à l’origine de l’ordonnance du président de la Cour du 12 octobre 2000, Grèce/Commission (C‑278/00 R, Rec. p. I‑8787, points 8, 16 et 18), dans laquelle l’invocation de « troubles sociaux très graves » a été écartée au motif que l’État membre concerné s’était borné à émettre des considérations générales dépourvues d’élément concret et n’avait fourni aucune indication quelconque quant à l’éventualité des graves événements allégués. En effet, contrairement au contexte de l’affaire C‑278/00 R, il est notoire que, en l’espèce, des perturbations de l’ordre public, telles que celles invoquées par la République hellénique comme conséquences prévisibles de la récupération imposée, se sont déjà produites dans des situations semblables, à savoir dans le contexte de mouvements contestataires dirigés contre les mesures d’austérité prises par les pouvoirs publics grecs depuis la crise économique.

51. Force est donc de constater que le cas d’espèce est caractérisé par des particularités établissant l’existence d’une urgence.

Thursday, September 20, 2012

Today in Luxembourg

No ECJ judgements this week, only AG opinions and General Court judgements:

AG Mazák proposed some guidance for the Greek Council of State in another gambling case. The AG’s discussion of the proportionality analysis that the national court is to undertake seems to reflect a certain skepticism about the likelihood that the state will prevail. Regarding the possibility of a transition period, should the existing system be found wanting, the AG is quite clear. That is a no go. Stanleybet et al. v. Ypourgos Oikonomias kai Oikonomikon

Predictably, AG Bot shot down  a blatantly discriminatory Polish rule for service of court documents. Cf. Regulation 1393/2007. Alder and Alder v. Orlowska and Orlowski

AG Sharpston had some issues with the question of whether the Greek Court of Auditors is a court in the sense of art. 267 TFEU for the purposes of the case at bar, but ultimately she concluded that it was. On the merits, she argued that you can’t let one category of (state) worker do union work “on the clock”, while counting the union work of another category as unpaid leave. Commissioner of the Elegktiko Sinedrio with responsibility for the Ministry of Culture and Tourism v. Audit Service of the Ministry of Culture and Tourism and Konstantinos Antonopoulos

AG Kokott had a similar admissibility problem with regard to the Bulgarian Commission for Protection against Discrimination. Kokott, too, ended up concluding that the case was admissible. On the merits, she argued that a Bulgarian law which implemented Directive 2000/43 only with respect to legally recognized rights was too narrow. Any unjustified disparate treatment that negatively affects an ethnic group is discrimination. And, as it turns out, putting Roma electricity meters at 7 m. above the street instead of 1.70 m. has a negative impact on them. Belov Cf. Recent Developments in European Consumer Law Blog and the UK Human Rights Blog

The General Court (Judge Wahl) agreed with the Commission that La Poste (FR)’s status as an EPIC (Établissement public à caractère industriel et commercial) and the manner said EPIC was governed created an implied unlimited guarantee of the state for the benefit of La Poste, which in turn meant that La Poste was receiving unlawful state aid. The Commission ordered that La Poste should be converted into a standard plc, and the General Court upheld that decision. France v. Commission (This is a topic that interests me greatly. I've blogged about it in the past, and I hope to do so again soon.)

Also important is the General Court’s decision in DEI v. Commission. The Commission had found an infringement of art. 106 TFEU, because the Greek government had given special and near-exclusive rights to DEI for the exploration and exploitation of lignite deposits. Cf. Commission Decision C(2008) 824. The General Court (Judge Kanninen) now held, however, that the Commission should have shown actual abuse of dominance, and that its failure to do so meant that the decision had to be annulled.


Last week, finally, there was another interesting competition case in the General Court – one that is still not available via Eur-Lex or in any other language than French – about the Commission’s discretion about which cases to pursue. In the end, the Court (Judge Kancheva) held that the Commission could have reasonably concluded that a sufficient Community Interest in enforcement was lacking. Protégé International v. Commission Cf. European Law Blog and Chillin’ Competition


UPDATE: Also interesting is this order of the President of the General Court, whereby he suspends the effect of a Commission decision requiring Greece to reclaim illegal state aid because if Greece tried to reclaim that aid people would riot. Cf. Verfassungsblog

Friday, September 14, 2012

This Week in Luxembourg

The Grand Chamber (per Judge Bay Larsen) dealt with some issues regarding the effect of Directives that have not yet entered into force in the environment law case of Nomarchiaki Aftodioikisi Aitoloakarnanias et al. v. Ipourgos Perivallontos, Khorotaxias kai Dimosion Ergon et al. The actual dispute concerned a partial diversion of the river Acheloos in upper Thessaly, which allegedly conflicted with Directive 2001/42 and Directive 92/43. Cf. European Law Blog


AG Trstenjak applied Directive 93/13 on unfair terms in consumer contracts to the energy sector. Interpreting the term “mandatory statutory or regulatory provisions” from art. 1(2) of the directive, she argues that this should also apply to provisions that the parties to a contract can opt out of. Combining that directive with art. 3(3) of Directive 2003/55, the AG finds that RWE’s practice of announcing price increases and then going ahead with them if no one complains is contrary to EU law. RWE v. Verbraucherzentrale Nordrhein Westfalen (NL, DE, FR) Cf. Recent Developments in European Consumer Law blog

AG Jääskinen considered the appeal in Switzerland v. Commission this week. While the General Court had bypassed the question of admissibility, relying on Council v. Boehringer and France v. Commission to reject the action on the easier merits aspect instead, the AG treats it head-on. He argues that Switzerland does not have the right to sue as a privileged applicant against an alleged violation of an EU-CH treaty, but that Switzerland’s action is nonetheless admissible because it is directly and individually concerned. On the merits he sides with the General Court. (The case was about Zürich airport, of course.)

AG Bot opined that in exceptional circumstances a Member State may decline to explain to someone why they’re being kicked out of the country despite the explicit rule of art. 30(2) of the Free Movement Directive. The problem is, of course, that if someone really is a serious threat to public security, the Member State will often prefer not to explain how they know this. The AG thinks the special advocate procedure is a good alternative. ZZ v. Secretary of State for the Home Department (NL, DE, FR)

AG Sharpston said: “Article 45 TFEU must be interpreted as meaning that a residence requirement such as that included in the skuldsaneringslagen (2006:548) (Law on debt relief) as a condition for obtaining debt relief constitutes a restriction on the freedom of movement of workers because it is liable to prevent or deter a worker from leaving Sweden to take up employment in another Member State.” Radziejewski v. Kronofogdemyndigheten i Stockholm


The General Court (per Judge Van der Woude) annulled the Commission’s state aid decision in Société nationale maritime Corse-Méditerranée. The Commission had found the bailout in question consistent with the Common Market, and now the General Court disagrees. Simply put, the Commission – and, by extension, France – were too generous in giving SNCM money to keep it upright rather than simply liquidating it and starting over. Corsica Ferries v. Commission (FR) Cf. Europolitics


Much to everyone’s disappointment, the ECtHR did not follow its precedent of Guzzardi v. Italy in Nada v. Switzerland. Instead of finding that being stuck in a 1,6 sq km exclave was a deprivation of liberty, the Court only held that it resulted in a violation of art. 8. Moreover, the whole UN Taliban sanctions list business was held to be a violation of art. 13, the right to an effective remedy. In case anyone was wondering, Guzzardi was “exiled” to Asinara, which is 56 sq km. Cf. EJIL: Talk!, the ECHR Blog and this blog post by Thobias Thienel

Thursday, September 06, 2012

This Week in Luxembourg

The ECJ is back with a vengeance. In Parliament v. Council (Schengen Borders Code), the Grand Chamber (Judge Von Danwitz) held that Council Decision 2010/252 was ultra vires, because it didn’t so much “implement” the SBC as legislate on essential elements of the subject-matter. The Court emphasized that in doing so the Council made political choices that potentially affected people’s fundamental rights, two factors that led it to conclude that this was a matter for the EU legislator instead. Cf. European Law Blog and my blog.

Another whopper has the Grand Chamber (Judge Bay Larsen) get involved in the debate about what to do about refugees who are persecuted for their actions, when those actions involve fundamental rights. Interpreting Directive 2004/83, the Court steers a middle course: not all interference with the freedom of religion is persecution, but a refugee cannot reasonably be expected to abstain from his religious practices. Germany v. Y and Z Cf. Verfassungsblog

In a fun bit of legal interpretation, the Grand Chamber (Judge Ilešič) held that art. 3(2) of the free movement directive does not create an automatic right of entry or residence for the people who qualify under that article, like dependants and non-spouse partners. They are entitled, however, to a careful examination of their case. Secretary of State for the Home Department v. Rahman et al. Cf. European Law Blog

Finally, the Grand Chamber (Judge Ó Caoimh) handed down a sequel to the 2009 Wolzenburg case. This time, the subject of the European Arrest Warrant is a Portuguese man living in France with his French wife. France, however, has only used the possibility of exception created by art. 4(6) Framework Decision for French nationals. The Court now holds that this is a violation of art. 18 TFEU and instructs the French court to evaluate the applicant’s connections with France. Lopes Da Silva Jorge


The judgement in Deutsches Weintor v. Land Rheinland-Pfalz is still just as dumb as the entire rest of this litigation. The questions, because they are too narrow, force the Court (Judge Malenovský) to accept the premise that “easily digestible” is a health claim under Regulation 1924/2006. The “hail Mary” attempt to bring the Charter into it also fails, although I do wonder why that argument didn’t mention free speech.

In Trade Agency Ltd v. Seramico Investments, the Court (Judge Tizzano) stood up for the rights of the defendant when it comes to the enforcement of in absentia judgements under Regulation 44/2000. The court that is asked to enforce such a judgement has jurisdiction to check whether the defendant had actually been served properly, and it may refuse to enforce an in absentia judgement where the original judgement is so obviously faulty that the defendant’s art. 47 Charter right to an effective appeal has been breached. Cf. UK Human Rights Blog

Regulation 617/2010 concerning the notification to the Commission of investment projects in energy infrastructure was enacted based on art. 337 TFEU (gathering information in general) and art. 187 Euratom (gathering information), but, according to the Court (Judge Ó Caoimh), the correct legal basis would have been art. 194 TFEU (energy). Somehow the latter is more “lex specialis” than the former. Parliament v. Council

If anyone wants to know how to define “pharmacological action”, which is part of the definition of “Medicinal product” in Directive 2001/83, the (lengthy and technical) answer (per Judge Borg Barthet) is in Chemische Frabrik Kreussler v. Sunstar Deutschland.

In the Fortis litigation, the Court (Judge Ilešič) held that the Dutch courts were allowed to summon the board members of Fortis to the Netherlands to testify. Regulation 1206/2001 simply means that the Dutch court was entitled, but not obligated, to have them heard in a (francophone) Belgian court instead. Lippens et al. v. Kortekaas et al.

“Article 15(1)(c) of Regulation No 44/2001 (…) must be interpreted as not requiring the contract between the consumer and the trader to be concluded at a distance.” That provision deals with consumer contracts specifically. Mühlleitner v. Yusufi and Yusufi Cf. Recent Developments in European Consumer Law Blog


In Railway Regulation law, AG Jääskinen published a whole series of opinions in infringement cases regarding Directive 91/440. He recommended that Spain (DE, FR) and Portugal should lose, that Hungary (DE, FR) should lose partially and that Germany and Austria should win. Once I’ve had the chance to look at these opinions more carefully, I’ll be able to say how the Commission managed to achieve only a 50% score.

According to AG Mengozzi, the Parliament’s most recent attempt to reduce the number of times it has to travel to Strasbourg is as flawed as the previous one. France v. Parliament (FR) Cf. EUObserver

AG Kokott does semantics and competition law: We know what it means for an agreement that restricts competition “by effect” to have an “appreciable” effect on competition, but what about a restriction “by object”? The AG explains that the Commission’s de minimis notice is not binding law, but that it is a guide for interpreting art. 101 TFEU. Discussing some case law, she gets no further than that such an effect must be shown somehow, but not (necessarily) by reference to market share thresholds. Expedia Inc. Cf. Kartellblog

AG Bot had an opinion on a judgement declining jurisdiction, which – he says – qualifies as a judgement in the sense of Regulation 44/2000. More interestingly, he says this judgement has res judicata for a 2nd court seized of the case, regardless of whether national law says so or not, with the exception of art. 35 of Regulation 44/2000. Gothaer Allgemeine Versicherung AG et al. v. Samskip

AG Jääskinen also has a procedural law mess concerning the appointment of experts in civil suits. Even though it is about a railway accident, I don’t think I want to know any more than that. ProRail v. Xpedys et al.

AG Trstenjak argued that a blanket ban on holding a sale without permission from the authorities is in violation of Directive 2005/29 on unfair b2c commercial practices, unless the authorities are allowed and indeed required to assess on a case to case basis whether there is actually anything unfair going on. Köck v. Schutzverband gegen unlauteren Wettbewerb (NL, DE, FR)

Wednesday, September 05, 2012

Implementing Powers

I was unexpectedly fascinated by what the European Court of Justice had to say in today's Schengen Border Code case. As others have also argued, the whole field of delegated acts/implementing powers is an oft-overlooked but vitally important area, with great potential for mischief. In January, I posted an introduction to the post-Lisbon state of affairs here. At the time, however, I was mostly concerned with the political wrangling between the European Parliament and the Council about where to draw the line between delegated lawmaking under art. 290 TFEU (which involves more power for the European Parliament) and implementing legislation under art. 291 TFEU (which favours the Commission and the Council). Today's case, however, is about an implementing decision that was taken under the old, pre-Lisbon comitology procedures, specifically the Regulatory Procedure with Scrutiny. (Cf. art. 5a of the Comitology Decision.) The question for the Court to answer was whether this was a valid exercise of the power delegated in art. 12 of the Schengen Border Code, or whether it was ultra vires.
Article 12
Border surveillance
1. The main purpose of border surveillance shall be to prevent unauthorised border crossings, to counter cross-border criminality and to take measures against persons who have crossed the border illegally.
2. The border guards shall use stationary or mobile units to carry out border surveillance.
That surveillance shall be carried out in such a way as to prevent and discourage persons from circumventing the checks at border crossing points.
3. Surveillance between border crossing points shall be carried out by border guards whose numbers and methods shall be adapted to existing or foreseen risks and threats. It shall involve frequent and sudden changes to surveillance periods, so that unauthorised border crossings are always at risk of being detected.
4. Surveillance shall be carried out by stationary or mobile units which perform their duties by patrolling or stationing themselves at places known or perceived to be sensitive, the aim of such surveillance being to apprehend individuals crossing the border illegally. Surveillance may also be carried out by technical means, including electronic means.
5. Additional rules governing surveillance may be adopted in accordance with the procedure referred to in Article 33(2).
Based on this last paragraph, the Commission proposed, and the Council ultimately adopted, Decision 2010/252. (Technically, what happened is that the Commission couldn't get the relevant Regulatory Committee to go along with its proposal, so it ended up on the plate of the Council. Parliament didn't object, and the Council ended up adopting it.)

According to its title, the purpose of the Decision is to "[supplement] the Schengen Borders Code as regards the surveillance of the sea external borders in the context of operational cooperation coordinated by the European Agency for the Management of Operational Cooperation at the External Borders of the Member States of the European Union". Specifically, the Decision "supplemented" the SBC with a number of rules and guidelines for sea border operations, such as what to do with people discovered at sea trying to make it to shore in vessels of varying levels of seaworthiness. (Think Libyan sea refugees.) Question: Is the Council entitled to do this in the exercise of its implementing powers under art. 12(5) SBC, or is this something that should have been done by the EU legislator, i.e. by Council and Parliament in collaboration?

When the Advocate-General's opinion came out, I was mostly amused by the argument that the Parliament's action was inadmissible because they didn't vote to block the adoption of this measure even though they could have. On the substance, I didn't look too closely, because I assumed it was a simple matter of tracking down the correct interpretation of art. 12(5) SBC. If the legislature, in enacting that provision, intended for this kind of implementing act to be enacted under comitology, it should generally be allowed. However, even at the time I already should have read more carefully. The AG, in recommending that the Decision should be annulled, relied heavily on the idea that the "essential elements" of a given subject-matter can only be enacted by the legislator, i.e. that implementing legislation cannot touch on such elements. This is an interesting idea, and one that apparently goes back to the 1970 case of Köster, but in the AG's opinion it came off a bit ill-defined. Apart from the question of whether the provision affects fundamental rights, no real definition was offered of what makes an element "essential".

Speaking with the mouth of the German judge Von Danwitz today, the Court gives much clearer guidance. To begin with, he reiterates the importance of fundamental rights as a factor in deciding whether something is an essential element of the subject-matter at hand:
77 [I]t is important to point out that provisions on conferring powers of public authority on border guards – such as the powers conferred in the contested decision, which include stopping persons apprehended, seizing vessels and conducting persons apprehended to a specific location – mean that the fundamental rights of the persons concerned may be interfered with to such an extent that the involvement of the European Union legislature is required.
More interestingly, he crafts something of a political question doctrine, except an entirely different one than the one most courts are used to. He tries to draw a line between decisions that are "political" and decisions that are "technocratic", leaving the former to the EU legislator.
76 [T]he adoption of rules on the conferral of enforcement powers on border guards, referred to in paragraphs 74 and 75 above, entails political choices falling within the responsibilities of the European Union legislature, in that it requires the conflicting interests at issue to be weighed up on the basis of a number of assessments. Depending on the political choices on the basis of which those rules are adopted, the powers of the border guards may vary significantly, and the exercise of those powers require authorisation, be an obligation or be prohibited, for example, in relation to applying enforcement measures, using force or conducting the persons apprehended to a specific location. In addition, where those powers concern the taking of measures against ships, their exercise is liable, depending on the scope of the powers, to interfere with the sovereign rights of third countries according to the flag flown by the ships concerned. Thus, the adoption of such rules constitutes a major development in the SBC system.
I think this is very interesting. Assuming it isn't taken too far, this is potentially a clear rule that can guide the political branches of the EU government in their quarrels over art. 290 and 291, at least when it comes to be boundary between those two articles and the general lawmaking power. Moreover, I think it is an excellent piece of judging. This paragraph does not attempt to craft constitutional principles wholesale. (Given that the EU legal order has always had, in practice, a very common law style, its judges sometimes forget the rule of art. 5 CC: "Il est défendu aux juges de prononcer par voie de disposition générale et réglementaire sur les causes qui leur sont soumises.") Instead, it sticks strictly to the case at bar, developing the existing case law (which is cited in par. 64 and discussed in the AG's opinion in par. 26-29, in particular) only as much as is necessary in order to resolve the dispute.

Yet, at the same time, this paragraph gives clear guidance to future judges. A matter is political to the extent that a) it requires a significant - and complex - weighing of interests, and b) there is a significant range of possible ways in which this can be done, i.e. a great number of policy options to choose from. As an additional factor, he offers the involvement of 3rd countries, which is sensible in this case but will probably not come up in most future cases.

Whenever I complain about the paucity of explanation given by judges (like here, in Dutch), this is roughly how I would like to see it. It is still not an academic treatise. Nor is it an attempt to use too much verbiage to hide the fact that the judge is not in the business of applying rules mechanically, but rather making a judgement call as he is paid to do. Instead, it cites case law and explains in  brief and clear language which considerations led the Court to decide as it did. Hopefully this case will help make the post-Lisbon Comitology 2.0 a little less scary.

Subsidiarity: Women Quotas


Now, I think we can all agree that having a quota for women on corporate boards is a monumentally bad idea. It is nobody's business but the shareholders' (and, in Germany, the workers') who sits on the board of a purely private company. Even if we were concerned about the glass ceiling, the kinds of women who sit on corporate boards are the last group to need the helping hand of the state. Quite the contrary. Introducing such a quota taints all female board members everywhere with the suspicion that they are not good enough, that they are only there to help the company meet its quota. It makes people wonder whether - to use an example from the European Commission - Commissioner Kroes has real skill, or whether she, like Commissioner Reading, is only there to hide the fact that the EU is still run by men. (I borrowed this picture of the 1991 Maastricht summit from Craig Willy. It goes to show that, to this day, the easiest way to make it to the top is to be born there. Too bad about the Queen's fashion sense.) 
Given how obvious this argument is, it is surprising that the UK seems to have based its opposition on the issue of subsidiarity instead. To be sure, this is not an implausible argument. It's just that, in Brussels, a subsidiarity argument generally only works in cases that are much more egregious still. I would have thought that it would have been much easier to win this argument on the merits. According to the Financial Times, my source for this story, the UK already has Sweden on its side. Bring in Ireland, and you're already half-way towards a blocking minority. The logical source for the rest of the votes is Eastern Europe, where they tend to be a touch more conservative and/or libertarian than in the West, which comes out in dossiers like this. Then again, you could also be cynical and combine the votes of the UK and Sweden with the votes of the six worst performers (i.e. everyone from Italy down in the graph on page 11 here, which surprisingly also includes Luxembourg). 
But since the British brought up subsidiarity, let's look at the draft proposal to see if that argument has merit. The Commission's discussion of subsidiarity covers more than an entire page, starting on page 8. It turns out that, while relatively long, this explanation fails spectacularly. 
Essentially, the Commission advances three arguments: 
1. Unless we do it at the European level, the Member States won't do enough.
That is fantastically dumb, quite possibly the dumbest explanation I have ever seen of why something satisfies the principle of subsidiarity. The fact that it is anticipated that the Member States will choose not to do as much as the Commission would like is in no way, shape or form an argument that suggests that action at the European level will work better. Giving the Member States the option of doing nothing is exactly the point of subsidiarity. In a perfect world, this would be an explanation of why this proposal would be shot down in the Council. As it is, it is a suggestion that the Members of the Council should use EU law to do a run-around their national legislatures and administrations. Neither of these points, however, is even remotely related to the question  of whether this law works better if it is enacted EU-wide. 
2. Level playing field, complications in business life and have a deterrent effect on companies' cross-border investments
These three all essentially come down to the same thing.  It's the standard go-to for the Commission in just about any dossier. The problem is that this is a company law rule, which means that every company only has to comply with whatever rule is in effect in their country of incorporation. There is no question here of one company being potentially subject to more than one rule. Nothing about this is particularly difficult to understand, companies deal with the fact that European company law is not harmonised every day. As for level playing field: Huh, weren't female board members supposed to help companies? More on that under argument 3. 
3. We need women on the boards of companies because this will promote economic growth that we badly need
This is wrong on two entirely different levels. First of all, I don't accept the premise. Assuming, for now, that companies are acting irrationally and promoting men to board-level positions instead of better qualified women, I refuse to believe that this effect is very significant, at a macro-economic level. No one person on the board of any company adds that much value. Not even big celebrity CEOs. (The more "celebrity" they are, the more they are subject to the rule of all autocrats: when they're good, they are very good, and when they're bad, they're very bad.) I don't see how all of this - by assumption - untapped potential can add more than a couple of basis points to economic growth. 
Apart from all of that, even if the economic growth at issue here were significant, it is still not enough to satisfy the principle of subsidiarity. It is not the task of the EU to promote economic growth by any means at its disposal. Instead, promoting economic growth is a task for the Member States, unless the EU can do a significantly better job of it. And, bearing in mind the previous points, why would the "growth spurt" created by Member State X introducing a women quota depend on whether Member State Y or Z also introduces such a quota?   

Taking all things together, I still don't think the UK chose wisely. In Brussels, a bit of hand-waiving along the lines of argument 2 usually gets you past subsidiarity, unless the Member States decide they don't want the proposal for other reasons. (Subsidiarity is a great argument to use if you want to get rid of a proposal for less politically correct reasons.) That doesn't change the fact, though, that the Commission did an appalling job here. Then again, maybe they did the best they could with what they were given. After all, I don't think I can think of any better reasons why this issue should be Europeanised.

Tuesday, September 04, 2012

NS Does a Jimmy Carr (3)

As explained yesterday, I managed to get all of Dutch politics in a frenzy last Saturday, by pointing out that the Dutch state railway company NS uses an Irish subsidiary to avoid Dutch taxes. However, on August 8th already, Farshad Bashir, a Member of Parliament for the SP (the Socialist Party) asked parliamentary questions about this affair to the Minister for Finance and the Minister for Transport. While we wait for the government to answer, it is useful to consider the answer given to fellow-SP MP Harry van Bommel about the exact same issue in 1999, when NS first established its Irish subsidiary. Here is my translation of the key question and the accompanying answer:
Question:
Have those members of the Supervisory Board that represent the State been consulted at all about [the plan to establish an Irish financing company for tax purposes]? If so, which position did they take? If not, is this consistent with the applicable legislation and with the articles of incorporation of NS?
Answer by the State Secretary for Finance, Willem Vermeend:
Under the Articles of Incorporation of NS plc, the Minister for Transport may appoint two supervisory board members. The appointment of one of these requires the consultation of the Minister for Finance. After discussions with the President of the Supervisory Board, it has been decided not to appoint anyone to this second post for the time being. The other government supervisory boardmembership is carried out by a non-civil servant. It should be emphasised that government supervisory board members do not sit on the board on behalf of the State; they are appointed by the government [fn 1: citing art. 2:158(12) Civil Code], but in the execution of their duties they focus on the interest of the enterprise [fn 2: citing art. 2:140(2) Civil Code]. This means that there is no right of instruction for the Minister, nor a duty on the part of the supervisory board member to inform. For this reason, it is impossible to give information about the decision making process in the Supervisory Board.
The two statutory provisions that are cited contain the heart of the problem, both at the time and still today. For this reason, I am sure that the answer to the questions posed by Mr. Bashir will come down to something along the same lines.

Problem 1:
Under art. 2:140(2) Civil Code, final sentence, the Supervisory Board promotes "the interest of the corporation and its enterprise". (For the non-lawyers, the corporation is the legal person, and the enterprise is the organisation. Not all corporations run an enterprise, and not all enterprises are incorporated, hence the distinction. Sometimes corporations even sell their enterprise to another corporation or individual. No problem. That simply means that the corporation is left with a board and a bag of money, and no further activities.)

Now, the interests of the corporation and its enterprise are not defined in the law, but they are assuredly distinct from the interest of the shareholder or shareholders. At least on paper, Dutch corporate law is an ode to the Rhineland model. The shareholders do not "own" the corporation, since a corporation is not something that can be owned. Instead, the shareholders have a claim, but so do the corporation's creditors, suppliers and employees. Without engaging in any detailed discussion of the jurisprudence and the literature we can say that the interests of the corporation can be equated with a reasonable balance between its continued healthy existence and the interests of the stakeholders, including the shareholders, of the company.

In this case, of course, the use of the Ireland-route did not benefit any stakeholder, nor was it significantly beneficial to the continued financial health of NS. Instead, the shareholder-stakeholder Dutch State was thrown under the bus to the tune of € 91 million and counting without any obvious countervailing benefit. The company is profitable, liquid and solvent no matter how it finances its rolling stock. However, neither I nor the Dutch State gets to make that point:

Problem 2:
Under art. 2:158 Civil Code, and more generally Part 6 of Title 4 of Book 2 Civil Code, the Supervisory Board of a "large" corporation is largely insulated from the reach of the shareholders. (Large is defined as having outstanding equity exceeding € 13 million in 2000-euros combined with having more than 100 employees.) Instead of having the Supervisory Board appointed at the General Meeting of Shareholders, as is the case for smaller corporations under art. 2:142 Civil Code, the Supervisory Board of a large corporation is essentially based on cooptation. The General Meeting has the option of rejecting a candidate for an open seat, a candidate chosen by the Supervisory Board, but only if it can marshal 50% of the votes representing at least a third of the company's equity against him/her. And even then, the worst that can happen is that the Supervisory Board has to propose another candidate. Absent litigation before the Corporation Chamber of the Court of Appeals in Amsterdam, no Supervisory Board of a large corporation can be forced to accept a member that it does not want.

It follows that, for a company like NS which has - in the legally relevant definition - € 1 billion in equity and has about 23.000 employees, not even the company's only shareholder gets to say what the company's goals should be. Nor does the State get to argue, as I just did, that avoiding Dutch taxation by financing its rolling stock through an Irish subsidiary hurts the shareholder without benefiting either the company or any other stakeholder.

Solution:
When I concluded my dissertation last year, I added one proposition that speaks exactly to these problems:
Neither liberalisation nor privatisation requires the use of private law instruments such as contracts and standard company types, such as standard-form public limited companies. Carefully designed public law instruments such as concessions and statutory companies should generally be preferred.
Now it should be said that at least for privatisation, i.e. for selling a formerly publicly-owned company to the private sector, the use of a standardised form of incorporation should generally be preferred. Using an ad hoc form will reduce the value of the company in the eyes of private shareholders due to the reduction in transparency if nothing else. Moreover, maintaining special government rights in an otherwise privatised company can get one in trouble with the European Commission, who have a long history of going after so-called golden shares as an infringement of the free movement of capital. (Whether I think that that is correct is a question for another day.)

In the case of NS, there is no question of it being privatised. There was some talk of this in the mid-1990s, but at the moment, and for the foreseeable future, there is absolutely no political support for this idea. Nor is the European Union legislator going to force this issue. At the European level, too, there would be virtually no votes for forcing Member States to privatise their railways. So for all practical purposes this is a question of liberalisation, not privatisation.

For liberalisation, my proposition is absolutely true. What I have in mind is what they call in France an Établissement public à caractère industriel et commercial (EPIC). An EPIC is a public entity that engages in industry or commerce (otherwise it would be an Établissement public à caractère administratif). The law does not, however, define in general terms how all EPICs are to be organised. Instead, there is a specific law or decree for each one, working from a general mould.

The best example, for present purposes, is the French railway company SNCF. It is established and organised by art. L2141-1 and following of the Code des Transports. In art. L2141-6, for example, we find how the board of SNCF is chosen, with further details in Decree 83-38, as amended.

The benefit of making such ad hoc rules is that, for every organisation that is placed at arm's length from the government, parliament and the crown can decide exactly how long that arm should be. In the case of NS, there are some unique provisions in its Articles of Incorporation, but the relevant corporate law provisions only allow so much creativity there. To my knowledge, there is nothing in the railway-specific legislation that overrules the general provisions of Book 2 Civil Code. And as a result, the Dutch government is stuck with a company that is disciplined by neither market nor hierarchy. The worst of both worlds.

Not even the concession can help here. Many things that used to be arranged hierarchically, by the Minister telling NS what to do, are now made mandatory by putting them in the concession for passenger transport on the core network. However, to use the power to proscribe how the service is to be carried out in order to determine how the rolling stock is to be financed would exceed the power of the Minister for Transport, I think. The purpose of the power to negotiate and adopt a concession is given in order to allow the Minister to engage in sensible, sustainable transport policy. To use this power to pursue a fiscal goal would be, in good Dutch law-French, détournement de pouvoir, i.e. abuse of right, and as such likely to be annulled by the administrative courts.

And so we all continue to do the same dance that we already did in 1999. The board of NS plays at being a private company, which it is not, the Minister for Finance loses out on several million euros each year, parliament is outraged at the "immoral behaviour" of NS while conveniently ignoring the fact that they are the ones that made the law this way, and the Irish get a little bit of extra bailout each year. Any number of people can do something about this, but nobody is going to.

Monday, September 03, 2012

NS does a Jimmy Carr (2)

OK, so that was quite a way to come back after my summer vacation. From this blog post here about tax evasion by the Dutch state railway company NS came this GeenStijl story, which in turn led to a ginormous article in De Volkskrant last Saturday, the short version of which is here. Result: all politicians scream bloody murder while simultaneously managing to mostly miss the point. More on that later. For now, I'd just like to bask in my glory a little longer and quote an English-language summary of the story:

From DutchNews.nl:

State-owned railway group NS uses Ireland to dodge Dutch taxes
Saturday 01 September 2012
The state-owned Dutch railway company NS has managed to cut its Dutch tax bill by at least €250m since 1999 by routing the cost of new trains through Ireland, the Volkskrant reported at the weekend.
The tax dodge means the treasury has lost out on income generated by a company it owns, the paper points out. The finance ministry, meanwhile, is said to be ‘unhappy’ about the arrangement, which it has been aware of from the beginning.
In effect, NS’s Irish subsidiary, NS Financial Services, has spent €1.7bn on new trains which it then rents to the NS in the Netherlands. None of the trains has ever been used on the Irish railways, the paper said.

Dividends
This allows the Dutch operation to avoid tax. In Ireland, railway companies have paid an average 9% tax on their profits in recent years. In the Netherlands, NS would have to pay 25% profit tax on the train rental. Some of the ‘missing’ cash does end up with the treasury in the form of dividends.
In a statement, the NS said the tax route had been developed to allow it to ‘better compete in the market’. Other large transport firms also use Ireland to reduce their tax liabilities and there is nothing illegal about this, the NS said.
The Volkskrant points out that there is effectively no competition on the Dutch railways and NS operates all intercity and most local train services.

Morals
Political party leaders were quick to react to the news. CDA leader Sybrand Buma told a Tros radio programme it showed a ‘lack of morals’. Labour leader Diederik Samsom said the NS had used a ‘bizarre construction which just is not right’, and an SP spokesman said the situation is ‘unacceptable’.
Caretaker tax minister Frans Weekers told the paper through a spokesman: 'Of course, we would rather have seen these activities take place in the Netherlands.'
Economist Martin Holterman, who is an expert on the Dutch railways, told the Volkskrant the NS is busy 'playing at being a company'. But the NS is not a company but a government service, he said. 

Just a few remarks, based on this version. (I haven't seen the original Volkskrant article yet myself.)

The tax rates are a bit iffy. The corporate documents we used did not allow us to calculate the effective rate of tax paid in Ireland. It is important to realise that the difference between Ireland and the Netherlands isn't just one of lower tax rates, but also of faster depreciation, which reduces the effective Irish tax rate for NS still further.

It is incorrect to say that there is no competition on the Dutch railways. NS has been given - without competition - the concession to operate trains on the "core network" until the mid-2020s, but there very much is competition for regional lines, where NS does not seem to be interested, and for the High-Speed connection between Amsterdam and Rotterdam. In the latter case, NS was so keen to win the tender that they increased their bid well above anything any expert deemed realistic, with the predictable result that they lost massive amounts of money and the concession had to be restructured. But that does not take away that there was competition for the award of this Fyra concession.

It is also incorrect to blame NS. They do their job as instructed: maximise profits. It is first and foremost the fault of the Finance Ministry that NS was not instructed and incentivised properly. More on this point in a later separate post.

As for my quote, I am not happy with the translation "government service". The key point is that NS is government-owned. So I would have written "a government entity" or something along these lines.

P.S. This article in the Irish Times this morning is better. However, they claim that I said that NS is "state-run", which is still not an ideal translation. The company is run independently, at arm's length from the state. That's exactly the problem, in this case.