It is trite in the arbitration milieu that the success enjoyed by Italy as a legal seat fades in comparison with the presence of Italian parties in international arbitration proceedings. Historically, the reasons for that were manifold, ranging from a (then) outdated arbitration law to a perceived mis-confidence of the Italian Courts towards arbitration.
Luckily enough, things appear to be changing.
The 2006 legislation reform has provided Italy with a detailed and comprehensive arbitration law, the arbitration culture has been steadily increasing among companies, and the arbitral community of practitioners has improved in number and quality. Also, some ad hoc amendments in the relevant legislation are expected in the next few months, in order to fill in the few remaining gaps with the most arbitration friendly jurisdictions (including the long-discussed power of the arbitrators to grant interim reliefs).
Far-reaching statistical data on the number and efficiency of arbitrations seated in Italy, though, are still missing. The report annually issued by the Milan Chamber of Arbitration (CAM) – the main arbitral institution of the country – shows a constant presence of foreign parties in the arbitrations administered by CAM and seated in Italy. It is not easy to know more than that.
The present contribution deals with a recent empirical study on setting aside proceedings in Italy, which is meant to deliver further useful insights on the matter.
The legal framework of an international arbitration seated in Italy
Differently from other jurisdictions, the Italian Arbitration Law is not embedded in a stand-alone act or statute but enshrined at the very end of the Code of Civil Procedure (CCP), in Sections 806-840 thereof.
In Italy, there exists no dichotomy between national and international arbitration. To the contrary, while Sections 806-838 CCP regulate in detail the procedure for arbitrations seated in Italy (be they national or international depending on the nationality of the parties and the features of the commercial relationship), Sections 839 and 840 CCP address the recognition and enforcement of foreign arbitral awards.
Therefore, any foreign party that is about to commence an arbitration seated in Italy might face the application of Sections 806-838 CCP exactly as it would happen in a domestic arbitration between two Italian parties. Said equal treatment, though, should not warrant much concern, as the Italian Arbitration Law – especially after the 2006 reform – clearly sets forth the majority of the overarching principles that parties to an international arbitration seated in an arbitration friendly jurisdiction would expect. To mention a few, the CCP acknowledges the separability and kompetenz-kompetenz doctrines, fosters a broad interpretation of the scope of the arbitration agreement, grants the parties the right to opt for institutional rules and endows the arbitral award with the same effects of a national judgement.
Focusing on the post-award stage, a party willing to enforce a domestic award needs to file an application before the Court of the district where the arbitration was seated, submitting the original of the award and the arbitration agreement. The Court, having checked the formal validity of the award (i.e., its existence as well as the existence of the underlying arbitration agreement), issues a decree declaring the enforceability of the award, which enables the award creditor to validly commence the enforcement procedure.
In case, however, either party believes that the award is to a certain extent flawed, supervening circumstances prove the ‘injustice’ of the award, or a third-party is unduly affected by its res judicata effects, the CCP enlists different procedures to file a request for correction, annulment, revocation or third-party opposition to the award, respectively.
The 2006 reform has intervened significantly also on those post-award instruments, with the double aim of (i) streamlining the procedures and (ii) fostering the stability of the arbitral award.
More specifically, the setting aside proceedings – which are the heart of the present contribution – may be summarized as follows. The annulment of the award may be requested on the basis of the specific grounds listed in Section 829 CCP only. Said provision provides twelve different motives to challenge the validity of the award, which are however quite narrow and could be largely traced back to the grounds for annulment embedded in Article 34 of the UNCITRAL Model Law.
Broadly, a request for annulment of the award may be grounded on (i) the invalidity of the arbitration agreement; (ii) the irregular constitution of the tribunal or incapacity of the arbitrators; (iii) the lack of jurisdiction or a decision exceeding the scope of the arbitration agreement; (iv) the absence of motivation, dispositive part or signature or contradictory provisions within the dispositive part of the award; (v) non-compliance of the issuance of the award with the applicable deadline; (vi) res judicata issues and (vii) due process violations.
Also, the award may be set aside in case the decision on the merits is mistaken as a matter of law, but only in case the parties to the arbitration agreement or the law had expressly envisaged said possibility. Lastly, the annulment may always be granted in case the award runs afoul of public policy.
The application may be filed before the Court of Appeal which district encompasses the place where the arbitration was seated. The time-limit for the filing depends on whether the award has been served or not. If so, the interested party has ninety days from the receipt of the service to file its brief. Otherwise, any party may challenge the award within a year from the last signature thereof.
The Court of Appeal decides on the application after hearing both parties, and may either grant the request (also partially, in case the part subject to annulment is severable from the others) or reject it. In case the setting aside is based on certain grounds, the Court may proceed to settle the dispute on the merits, unless otherwise agreed by the parties. It is worth stressing, though, that the presumption is reversed in case either party was resident abroad at the time of the execution of the arbitration agreement, so that the Court of Appeal will be prevented from deciding the merits of the dispute in the absence of any agreement to that end.
At last, the decision of the Court of Appeal may in turn be challenged before the Supreme Court of Cassation on very limited grounds.
The scope and purpose of the statistical research
Having briefly depicted the relevant legislative framework, the present study stems from the analysis of over a thousand judgements deciding annulment applications filed before the Courts of Appeal of Bologna, Milan, Naples, Turin and Venice. The time-frame considered, which slightly varied due to the limited availability of older judgements in electronic format, always encompassed – for consistency purposes – at least five years in the period from 2012 to 2018.
The examination of the judgements offered a deep insight into the Italian arbitration world, from two different standpoints. On the one side, it was possible to retrace many information on the arbitration that led to the challenged award, from the gender and profession of the arbitrators to the nature of the underlying dispute.
On the other side, there was the chance to examine in detail the development and peculiarities of the setting aside proceedings, understanding – inter alia – the frequency and ‘success rate’ of the grounds for nullity mostly invoked by the parties.
The reason behind the research were manifold. Firstly, it seemed a good way to check whether the Italian judiciary system has definitely shrugged off that (in many instances only perceived) diffidence towards arbitration. Secondly, the window opened by the setting aside proceedings on the arbitration provided enlightening data on a world that is normally covered by confidentiality. Thirdly, the study provided parties and counsels with multiple hints on how judges deal with setting aside requests, which should promote a more efficient use of said procedural instrument. Lastly, the research was meant to contribute to the sharing and fostering of the arbitration culture among all the different stakeholders (including judges), which is key to the growth of Italy as an arbitration-friendly place.
While it is too early to tell whether those aims will be fulfilled, the results of the research definitely give confidence for the continued development of Italy as a valid seat for an international arbitration.
The most significant results
As anticipated, the data obtained from the research may be roughly divided in two broad categories. Namely, those relating to the underlying arbitration and those relating more specifically to the setting aside proceedings. A brief overview of the main results follows accordingly.
The underlying arbitration
The type of arbitration
Starting from the type of arbitration, the analysis showed that the arbitrations seated in Italy are still predominantly ad hoc, with only 14 per cent of the cases being administered by an institution. The notable exception, probably explained by the presence of the CAM, is Milan, where one out of four cases are institutional proceedings.
As to the source of the tribunal’s decision-making, pursuant to Section 822 CCP the arbitrators should settle the dispute on the basis of the law, unless the parties have provided for their duty to issue an ex aequo et bono decision. Such default preference appears clearly in the judgements examined, which dealt with the possible annulment of ex aequo et bono awards in only 8 per cent of the cases.
Shifting the focus on the parties to the setting aside proceedings (and hence to the prior arbitrations), the research confirms arbitration as a reliable dispute resolution mechanism for a variety of stakeholders:
As to the origin of those parties, the data show that in Italy arbitration is widely used also for settling domestic disputes, with the majority of the cases brought to the scrutiny of the State courts being between Italian parties. In fact, only 6 per cent of the cases are truly international, with only the Milan Court of Appeal reaching double digits. For the purposes of this study, only the registered office of the legal entity involved in the arbitration has been taken into consideration to determine the nationality of the arbitration. It often happens, however, that the legal entity party to the arbitration is Italian but fully controlled by a foreign mother company, so that the dispute is only formally a domestic one.
At last, multi-party proceedings took place in one case out of four, corroborating the wide-spread understanding of arbitration as a viable mechanism to settle complex disputes.
On the composition of the tribunal, the default legislative preference for a three-membered tribunal reflects in practice, with the presence of a sole arbitrator in only 21 per cent of cases on a national scale. The number, though, significantly varies among the different Court of Appeals, ranging from the 8 per cent of Naples to the 40 per cent of Florence.
Cross-checking the data, it was possible to investigate the much-debated topic of whether, and if so to what extent, the composition of the tribunal affects the stability of the arbitral award. The results, however, did not appear decisive, as the percentage of annulment requests granted did not vary significantly depending on the composition of the tribunal.
To the contrary, it emerged an established tendency of the parties to withdraw their setting aside applications filed against awards rendered by sole arbitrators. And that could be likely explained with the usually smaller value in dispute in those cases, which may prompt the parties to find an out-of-court settlement in lieu of continuing the setting aside proceedings.
Touching upon another hot topic in the arbitration world, data were collected on the gender and profession of practitioners settling Italian seated arbitrations.
As to the former, Italy seems in line with the despicable international practice of having an insufficient (but at least rising) number of women sitting as arbitrators. Notably, the adjusted percentage on a national level stopped at around 10 per cent, with some positive exceptions in the arbitrations seated in Venice and Florence. Be that as it may, much has to be done in relation to gender equality in arbitration. Hopefully, the growing number of institutional appointments of female arbitrators will help overcoming said rusty stereotype.
As to the latter, the researchers found as follows:
Adjusting the percentage to take into account the 43 per cent of judgements where the information was not provided, non-legally qualified professionals accounted for around 15 per cent of the arbitrators who adjudged Italian-seated arbitrations. The number – driven by the numerous construction cases as well as the controversies within professional associations – definitely pays tribute to one of the oft-cited advantages of arbitration, namely the possibility to have private judges with a specific expertise.
The underlying dispute
The last elements worthy of mention are the subject matter and value of the controversies settled by arbitration. Regarding the matter in dispute, data appear quite homogeneous and confirm the parties’ reliance on arbitration as a mechanism to effectively settle dispute in different fields:
As to the value of the underlying arbitration, this is probably the less reliable data emerging from the study, due to the high number of judgements bearing no such information and the presence of multiple disputes with an undetermined value, e.g. challenges of a shareholders’ resolution in corporate disputes. In any event, the average value of the arbitrations subject to annulment proceedings was equal to EUR 1.883.011,83, with a significant discrepancy between the cities with a bigger case-load (e.g. Milan, avg. value EUR 4.240.145,11) and the ones with less arbitrations (e.g., Venice, avg. value EUR 515.733,03). Even bearing in mind the caveat on the reliability of the data, the parties’ preference to opt for arbitration over domestic litigation when the stakes are high appears settled.
The setting aside proceedings
The grounds for annulment
Against the above-described legislative framework, the ensuing pie chart reflects the frequency of use of the different grounds for nullity as invoked by the parties and their counsels.
As apparent, although the 2006 reform reversed the presumption on the possibility to challenge the award on the basis of an error of law, there is still a tendency among parties to have a try with such ground, invoked in almost 20 per cent of the judges examined. Two other numbers are worth citing.
Firstly, due process complaints – in the sub-categories of evidentiary issues, issues relating to terms and more general due process issues – accounted for 13 per cent of the cases. However, it will be clear from the analysis of the ‘success rate’ of this ground that said number reflects a yet diffuse malpractice of raising due process issues at the post-award stage only as a last resort remedy.
Secondly, challenges based on the perimeter of the decision in the arbitral award (be it allegedly out of the scope of the arbitration agreement or infra petita) were raised in 22 per cent of the cases under scrutiny.
The outcome of the annulment application
Going to the heart of the study, the data examined enable to conclude that awards issued by Italian-seated tribunals are reasonably stable, and the annulment of the award is granted only under very peculiar circumstances.
Notably, the adjusted rate of rejection of setting aside requests on a national level rises at 84 per cent, with an 11.5 per cent of cases where the award was annulled and a 4.5 per cent of partial annulments. Also, the vast majority of partial annulments concerned quite minor issues within the award (e.g., the allocation of legal fees) and did not affect the main heads of claim adjudged by the arbitrators.
On a local level, the adjusted rate of rejection ranges from 82 per cent to 88 per cent in all but one of the Courts of Appeal under scrutiny, thereby confirming that the judges’ deference to the finding of the arbitrators is quite homogeneous in the main Italian cities.
The ‘success rate’ of the grounds
Cross-checking the data on the frequency of use and the outcome of the proceedings, it was possible to have an idea of which grounds of nullity are more likely to succeed before the Italian judges. While the most frequently invoked grounds led very frequently to a rejection of the application, there are instead other grounds that seem to be more appealing to the judges ruling on the annulment requests.
To mention a few examples, while applications based on alleged due process violations failed in almost 98 per cent of the cases, challenges based on the invalidity of the arbitration agreement were granted in 16 per cent of the circumstances. Also, awards allegedly in contrast with prior awards or judgements were annulled in 15 per cent of the cases, whereas only one request out of twenty was granted on the basis of errors of law.
The tentative takeaway is that the grounds for nullity that even prima facie appear as more objective and ‘neutral’ meet more success than the broader ones, which are generally discarded by the judges.
The average length of the setting aside proceedings
In cauda venenum, Latins said (literally, ‘the poison is in the tail’). A recurring criticism still made to Italy by foreign investors is the length of the judicial proceedings. A significant work has been done in the recent years and the upcoming reform of the judicial system, which is currently being drafted by the Italian Government pursuant to the Parliament’s approval of the enabling act, is expected to cause an important systemic evolution to the judicial system (with a positive impact also on arbitration). The study, which inevitably offers a retrospective picture, shows quite diverging results that are worth of consideration, and once more demonstrate that the determination of the seat of an arbitration is an important choice in many respects.
The length in months of the setting aside proceedings varies significantly from one Court of Appeal to the other, and presents room for improvement, especially in judicial seats such as Bologna and Venice. The Courts of Appeal of Turin and Florence, managing to conclude the proceedings in less or around two years, are notable exceptions, with a duration of the proceedings which is in line with that of many other jurisdictions known as arbitration friendly. Milan, with a significantly higher number of cases to handle, sees its setting aside cases concluded in slightly less than three years, a data which places itself somewhat in the average of the Italian Courts of Appeal but that certainly has to be improved.
At the end of this small journey in the Italian-seated arbitrations and the related setting aside proceedings, a few closing remarks appear due.
Overall, the analysis of case-law suggests that Italian judges have a positive view of arbitration and tend to respect the finality of the arbitral award. Indeed, even more so after reading each judgement, it can be said that the annulment of the award is declared only when one of the narrow grounds listed in Section 829 CCP is fully met, and avoiding to do so would render the judicial control moot.
To the contrary, in several seats there is still a long way to go in relation to the length of the setting aside proceedings. Inevitably, a fecund growth of international arbitration in Italy will have to deal with the issue. To that end, the reform of the judicial system that is currently in the hands of the Italian Government will help the system moving further ahead towards efficiency and celerity.
With regard to the partial window on the underlying arbitrations, the resulting picture is quite comforting. Arbitrations seated in Italy are more and more used by foreign parties, appear to be capable of effectively adjudicating complex and different issues, and are decided by arbitral tribunals with different backgrounds and (rising) gender equality, especially in institutional arbitrations.
More generally, the present statistical study integrates in a context of a growing arbitration culture and openness towards international arbitration in Italy. This and other contributions are needed in order to foster the sharing of knowledge and data that are key to the development of Italy as a mature jurisdiction to host international arbitrations.
According to the enabling act approved by the Italian Parliament, the reform of the judicial system will most likely boost the role of arbitral tribunals seated in Italy, for instance giving them the power (that they currently do not have) to issue interim reliefs. Such envisaged changes go in the right direction towards a modern arbitral environment, which will need to be backed up by a robust judiciary system.
 Suffice it to mention that, pursuant to the ICC Dispute Resolution 2020 Statistics, Italy shares the fourth place (alongside France) in the ranking of the most frequent nationalities in ICC arbitrations, sitting less than 0.5 per cent behind the third place held by Spain. Quite to the contrary, Italy has 13 jurisdictions ahead in the ranking of the jurisdictions most frequently selected as place of the arbitration.
 See Sections 808.2, 808-quarter, 832, 824-bis CCP, respectively.
 M. Benedettelli, INTERNATIONAL ARBITRATION IN ITALY, Kluwer Law International, 2020, p. 400.
 To be thorough, whenever a judgement did not specify the nature of the underlying proceedings, the case was accounted as ad hoc in light of the judges’ established practice to expressly mention the administered nature of the arbitration only.
 Once again, the data provided by CAM are even more encouraging, with 22% of the arbitrations administered by the Institution in 2021 being international (data yet to be published on CAM’s website but provided by the Institution).
 Section 809 CCP. Conversely, Art. 14 of the CAM Rules provides for a default choice in favor of a sole arbitrator, subject to the Arbitral Council’s discretion to refer the dispute to a three-member panel in light of its value and/or complexity.
 See, e.g., ICC Dispute Resolution 2020 Statistics, p. 5.
 The percentage was ‘adjusted’ considering that in almost half of the judgements examined it was not possible to discern either the gender and the profession of the arbitrators, due to some judges’ tendency not to indulge on the recollection of the arbitration that led to the challenged award.
 See CAM Annual Report on Arbitration, p. 8 (available at https://www.camera-arbitrale.it/upload/documenti/statistiche/statistiche%20arbitrato%202020.pdf).
 The rate was adjusted to take into account the 13 per cent of cases where the court proceedings did not end up with a judgement, usually due to the parties’ decision to find an out-of-court settlement and withdraw from the setting aside proceedings.
 It is fair to note that said conclusion is influenced by the high percentage of rejection of setting aside applications, which naturally has a bigger impact on the mostly invoked grounds.