It must be an arbitrator’s nightmare: Imagine a high-stake arbitration that goes on for years, the entire distance, including witness hearings and expert evidence, only for the final award to be set aside on procedural grounds. And this is exactly what the Frankfurt Court of Appeals (Oberlandesgericht Frankfurt) did in a judgment in February 2011: The post-M&A dispute between GEA and Flex-n-Gate began in October 2004. There were, for some time, parallel court proceedings in various jurisdictions. An interim award was made in 2006. The final award, rendered in 2010, was for EUR 210,658,362. In October 2012, the Frankfurt judgment which reversed it has been upheld by the Federal Supreme Court (Bundesgerichtshof). What had happened?

In 2004, German technology group GEA sold one of its subsidiaries to U.S.-based Flex-n-Gate. The SPA, governed by German law, provided for DIS arbitration. The transaction was never consummated. The dispute that followed from that led to arbitral proceedings being commenced. The arbitral tribunal split the proceedings into a liability phase and into a quantum phase. During the liability phase, several hearings took place and witnesses were heard. On that basis, in September 2006, a partial award was issued. It found respondent liable, and dismissed respondent’s counter-claim.

In October 2006, the arbitral tribunal turned to the quantum phase. It sent a draft procedural order to the parties, inviting the parties to comment on the draft and suggest changes. Taking into account the parties’ comments, the final version of the procedural order defined how the quantum phase was to be conducted. In particular, under the order, the parties were to disclose the documents that they had made available to their own experts. In the preamble, the order said that it recorded the agreement between the parties regarding the conduct of the quantum phase. Separately, the terms of reference had stated that the expert appointed by the tribunal was to carry out his own calculations, and should not rely on the calculations performed by claimant’s expert.

During the quantum phase, the claimant did not disclose all of the documents set out in the procedural order, nor did the expert appointed by the panel carry out his own calculations. Nevertheless, a final award in favour of the claimant was made in March 2010. Respondent then applied to the Frankfurt Court of Appeals to set the arbitral award aside. This application was based on various arguments, of which one ultimately was successful:

Respondent argued that the procedural order, as a result of having been agreed between the parties, constituted an agreement between the parties within the meaning in Sec. 1059 German Code of Civil Procedure (ZPO). Sec. 1059 ZPO provides that an arbitration award may be reversed, inter alia, if “the arbitration proceedings did not correspond to (…) an admissible agreement between the parties.” In respondent’s opinion, the arbitral tribunal was bound by the agreements between the parties as recorded in the procedural order. The tribunal had no discretion to unilaterally deviate from the procedural order during the quantum phase. As the procedural order had not been fully complied with, the arbitral proceedings did not correspond to the agreement between the parties and hence, the award had to be set aside.

The Frankfurt Court of Appeals (Oberlandesgericht) was persuaded by that argument and did set the award aside on that basis. The Frankfurt court looked through the formal nature of the procedural order and held, on the basis of the evidence before it, that in substance the order was an agreement between the parties. The Court gave particular weight to the fact that the wording had been negotiated in detail. In reaching this conclusion, it did take, inter alia, evidence on the content of a telephone conference between the parties’ lawyers and the tribunal. The Court applied the same analysis to the terms of reference. Claimant’s application for legal review (Rechtsbeschwerde) to the Federal Supreme Court (Bundesgerichtshof) was dismissed as inadmissible on procedural grounds, without stating any reasons (file no. III ZB8/11). The judgment of the Frankfurt Court of Appeals thus has become final.

The Frankfurt judgment was widely discussed in the German arbitration community, for instance at the DIS Autumn Meeting in October 2012, when the topic was on the agenda. I believe there is a wide-spread consensus that the approach taken by the Flex-n-Gate arbitral tribunal was very much in line with current international practice and the Frankfurt court decision came quite as a surprise. It was criticized for over-emphasizing the concept of party autonomy, as unreasonably interfering with the arbitration process and as sending the wrong signals to arbitrators by discouraging co-operative approaches to the conduct of arbitration. But Flex-n-Gate clearly poses a challenge to that practice for any arbitration with its seat in Germany. What some commentators have labelled “hybrid procedures” have, all of a sudden, become risky and may back-fire. Going forward, arbitral tribunals will need to be very clear in their communication, and distinguish whether they are in a co-operative modus operandi, or whether they are issuing procedural orders in their arbitral capacity. In the first instance, arbitrators will need the parties’ consent if they want to revise the game plan and deviate from the order, whereas in the second alternative, arbitrators will retain the liberty to amend existing procedural orders in their sole discretion.


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

Profile Navigator and Relationship Indicator
Access 17,000+ data-driven profiles of arbitrators, expert witnesses, and counsels, derived from Kluwer Arbitration's comprehensive collection of international cases and awards and appointment data of leading arbitral institutions, to uncover potential conflicts of interest.

Learn how Kluwer Arbitration can support you.

Kluwer Arbitration
This page as PDF


  1. I am amazed at the final paragraph “I believe there is a wide-spread consensus that the approach taken by the Flex-n-Gate arbitral tribunal was very much in line with current international practice and the Frankfurt court decision came quite as a surprise.”

    The process is a matter of party consensus and – OF COURSE – party autonomy is paramount and any legal agreement between the parties must be honoured or the tribunal should withdraw.

    1. Hi, actually, the litigation is not a matter of consensus, the only consesus in arbitration is the arbitration clause/agreement. When the arbitration starts, the arbitral tribunal/sole arbitrator is called to apply the arbitration rules through its procedural orders and finaly through the award, in a compulsory way, and even against the parties’ consent, if the particular circumstances so require. There is no significant difference between a court order and an arbitration procedural order. Therefore, if in the above case, Frankfurt Court of Appeals had been called to rule on a decision issued by an inferior court, on an identical issue, would it have ruled in a similar way, by stating that the court order was the will of the parties? I really do not think so.

    2. I was present at the DIS conference and there certainly wss no such consensus. I agree to what was said downthread. The blog misses important langauge of the decision: This tribunal had expressly asked for a parties agreement and failed to honor it in the course of the proceedings. And acted against the most important principle of arbitration, party autonomy. If you read German, the decision is public on

      1. Thanks for the comments. The link to the judgment had already been included in the post. – @Thomas Weimann: Would you agree to say that it was a widely-held opinion, rather than consensus, and would you be prepared to disclose your personal involvement in the matter?

        1. My personal involvement In the matter is obviously public domain. No, I would not agree that it was a “widely held opinion”. I remember one clear statement against the Frankfurt decision from claimant’s Supreme Court counsel and one statement from the mangaging partner of a huge domestic German law firm. The broad consensus was, that the facts given (and summarised a bit too short in your blog) did hardly allow any other decision and the tribunal made mistakes which were hard to understand.

  2. Nach dem Wortlaut der inzwischen veröffentlichten Frankfurter Entscheidung ist Ihre Sachverhaltsdarstellung deutlich verkürzt.

    Dort heisst es nämlich auch: “Der Vorsitzende des Schiedsgerichts wies in diesem Zusammenhang darauf hin, dass das Schiedsgericht die Beweisaufnahme streng nach den vorgeschriebenen DIS-Regeln bzw. der ZPO durchführen werde, sofern die Parteien keine abweichenden Regelungen vereinbaren würden”.

    Erst dann kommt es zu dem Text, der eingeleitet wird: “Das Schiedsgericht hält des Weiteren die Vereinbarung der Parteien zum Inhalt der weiteren Schriftsätze fest” und dann folgt die Vereinbarung.

    Wenn sich ein Schiedsgericht mit der im Tatbestand dokumentierten Vorgeschichte nicht daran hält, dann ist das kein “Nightmare”, sondern schlichtweg “grobe Unvernunft”.

    1. According to the lanugage of the PO as it is reported in the decision, the procedural order did not “constitute” the agreement of the parties but “documented” the agreement of the parties. And you cannot change parties’ agreements without the parties.

  3. Paul Bennett Marrow from NY has shared this comment with me, and has given me permission to reproduce it here:

    “Interesting case. But I think the German Court was right. Your arbitration statute and the ruling of the Court are in agreement with what we see here in the US.

    In the US most arbitrations are subject to the F.A.A. Section 10(4) allows a court to “vacate” an award when it finds that an arbitrator has exceeded his/her/their powers as setforth in the arbitration clause itself. There is a study by the American Bar Association’s section on Dispute Resolution that found that while having an award vacated is relatively rare, the claim most likely to succeed is one brought under Section 10(4). Why? Because arbitration is a function of contract.The hearing is theirs and they have the right to structure a hearing in any manner they see fit so long as the structure isn’t otherwise offensive to the law. Judge Richard Rosner once famously noted: “…short of authorizing trial by battle or ordeal, or more doubtfully, by a panel of three monkeys, parties can stipulate to whatever procedures they want to govern the arbitration of their disputes; parties are as free to specify idiosyncratic terms of arbitration as they are to specify any other terms in their contract.” Baravati v. Josephthal, Lyon & Ross, 28 F.3d 704, 709 (7th Cir. 1994).

    Arbitrators aren’t judges. They have powers that are limited by the wishes of the parties. That is the base line. The fact that an award was issued after years of testimony etc. makes no difference.Once the arbitrator falls of the wagon and goes off on his/her/their own the damage is done.

    Moral of the story: Whether in Germany or in the US (and in most other jurisdictions as well) if your an arbitrator don’t exceed your power. People are going to punish you for doing that even if you were acting in good faith. And if you do go off and then have the award vacated, don’t expect parties to pick you again anytime soon in the future.”

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.