Corruption Evidence in Arbitration: From Red Flags to Expert Reports

More voices are noting the need for arbitrators to be aware of and actively deal with signs of corruption. But how best to find and present evidence of corruption in arbitration proceedings?

The circumstances surrounding corruption can be complicated. A targeted population of transactions over a focused time period is important, together with a robust methodology for identifying and describing potentially problematic business transactions or red flags.

Consider red flags collectively  

The concept of red flags has developed to encompass a wide range of circumstances that may indicate an actor wilfully attempted to obtain or retain business through a transfer of value or promise of transfer of value. These include unreasonably large discounts to third-party distributors and the use of third parties related to (or closely associated with) foreign officials (Resource Guide to the US FCPA, 2020). These indicia are best considered collectively and not only as single occurrences.

For example, in November 2020, in the Sorelec v. Libya case, the Paris Court of Appeal set aside a French construction company’s US$452 million International Chamber of Commerce award against Libya after finding it was based on a fraudulent settlement agreement and, as such, incompatible with international public policy. In September 2022, the French Court of Cassation upheld the annulment decision thereby confirming the position it had taken on the Belokon v. Kyrgyzstan case a few months earlier. [4]

In the Sorelec case, and in line with the Paris Court of Appeal’s doctrine, the standard of proof applied was the red flags test, an approach also put forward by FCPA practice. This requires the party alleging the acts of corruption to prove that the acts are sufficiently ‘serious, specific and consistent’ to carry a presumption of culpability. If the party alleging corruption is successful in proving so, the burden of proof falls on the opposing party. The Court of Appeal identified four circumstantial elements that convinced the judges of the veracity of the allegations:

• the chaotic and ‘corruption-friendly’ situation in Libya at the time of the conclusion of the disputed settlement;
• the peculiar procedure followed for the conclusion of the settlement, in particular the absence of certain administrative authorisations;
• the hastiness with which the settlement was concluded; and
• the peculiar terms of the settlement, in particular the absence of reciprocal concessions, which was contrary to Libya’s interest, thereby proving that the minister in charge of the settlement had received a bribe.

However, not all red flags are created equal. As Lucinda Low observed, “some red flags . . . especially those that are transaction-specific, are more probative than others; conversely, general red flags such as country risk have very limited probative value”. [5]

Therefore, when defining legal strategy, it is important for counsel to understand and weigh the factors appropriately.

Instructing forensic accounting experts

The role of the forensic accounting expert in this context can cover a variety of areas: expectations for best practice in the international business environment (e.g., regarding a rigorous risk assessment, a robust internal control environment or an effective ethics and compliance programme); the review of specific transactions for red flags, including tracing cash flows or identifying and evidencing conflicts of interest.

It is important to keep in mind that red flags must be factual and specific, and considered together rather than in isolation

It is important to keep in mind that red flags must be factual and specific, and considered together rather than in isolation. Experts should identify these indicia, list them out and describe how they stack up in the overall argument. For example, a 3 per cent remuneration for an agent operating in Western Europe with an established business may indicate a low-level of risk. However, further red flags – such as operating in a country with historically high levels of corruption, maintaining a banking location in a tax haven, or the absence of internet presence or credentials in the relevant industry – indicate a pattern or more compelling risk of corruption.

Examples of red flags, some of which are identified in the Basel Toolkit for Arbitrators, include the following:

• unusual payment methods (e.g., cash, gifts or other indirect payments)
• use of third-party intermediaries with little experience or that carry reputa- tional risk (e.g., a third-party intermediary investigated or fined by national or foreign regulators or authorities)
• use of offshore or shell companies with an opaque ownership structure
• circumvention of due diligence processes in selecting vendors
• lack of details around and evidence of services rendered and billed
• high or unusual amounts of payment not benchmarked to activity levels or market prices
• non-compliance with local tendering  rules  (e.g.,  unusual  speed  of  bidding process)
• little or no online presence
• the absence of a well-designed and active compliance programme

In the context of international arbitration, instructing an expert to identify and interpret red flags is likely to be focused on specific transactions that form part of the arrangements giving rise to the underlying dispute (which differs from an internal or regulatory investigation, for example, which may have a wider risk- based scope). The scope of the expert’s instructions will then guide the extent to which data will be collected to be reviewed for relevant red flags. Access to financial data, transaction and banking documents will be key to establishing the existence of red flags and potential corruption.

Scoping data collection

Obtaining data and conducting the review could well become iterative and, subsequently, a lengthy process. Aspects of evidence such as data type, data location and custodian will impact the expert’s methodology. As such, it is important for counsel to coordinate with the expert on the development of a sound approach tailored to the facts and the parties, and to ensure that the overall arbitration plan provides for disclosure requests for both sides to avoid last-minute unmanageable efforts.

Data is generally described as either structured or unstructured. Structured data relates to data that is highly organised (for example, enterprise resource planning (ERP) system data) or accounting data stored in tables with defined fields. While it is simpler to search and analyse this data, where there are multiple sources of the data work may be required to normalise and prepare it for efficient query and interpretation. Unstructured data relates to information from a variety of formats, such as emails, contracts, invoices or bank statements. This data usually requires more work to collect, process and understand.

Due to its complexity, data collection is often a time-consuming and potentially costly venture. Therefore, it is important to approach it only after gaining a solid understanding of who the data custodians are, the time periods covered by the data, and where and how the data is stored. Depending on the sophistication of where and from whom data is being collected, the difficulty of obtaining an accurate and complete set of data may increase. Collection can involve large volumes, which can take time and may well be met with resistance. This is the case when collecting structured data from an ERP system over multiple years.

For all these reasons, adopting a proportional and phased approach to data requests and collection is key to controlling the time and cost of the analysis. For example, adequately scoping the relevant period before collecting accounting data will save unnecessary time spent processing, reconciling and analysing data. Another example in the case of large companies with multiple subsidiaries and affiliates is to start with the data relating to the most obvious legal entity, which in most cases is the one involved in the contract under dispute. As the data analysis and concurrent legal document review progress, other legal entities could be included as necessary based on the likelihood of their involvement.  

The parties and their counsel should be ready to defend their scoping approach before the tribunal. Consistency and justifiability of the approach throughout the case is key.  

The same principle goes for the collection of unstructured data, for which identification of the relevant custodians, location and imaging or copying of documents should be performed as early as possible to avoid destruction of records. Reviews of structured and unstructured data feed into each other in an iterative manner until relevant accounting entries for transactions of interest and supporting documentation, such as contracts, invoices, approvals and email communications, are identified.

Practical tips for a compelling expert report

After the challenge of collecting and analysing data, the culmination of efforts by counsel and experts is in reporting the results – either in writing or during oral testimony. Throughout the build-up to reporting, the experts should never lose sight of their duty to support the tribunal in understanding the issues at hand. Therefore, the experts’ main responsibility is to explain their reasoning and conclusions as clearly and practically as possible.

Experts should set the stage for their analyses and conclusions by clearly laying out their instructions, relevant context and circumstances (including region and industry), the assumptions and documents on which they have relied, the methodology adopted to assess indicators of corruption, and the type of evidence they obtained and how they obtained it.  

Equally as important, the experts should be upfront about any relevant evidence that may have existed but could not be obtained, or about data that was discarded and why. There may be justifiable reasons for the absence of data, such as the lack of consistency, completeness or accuracy of data sets, or statutory document retention periods. However, it may be that the absence of data and documents in disclosure raises the prospect that information that is relevant and material to the matter has not been disclosed, and in these circumstances, counsel may consider whether the facts could suggest adverse inference.

Remember: red flags are not smoking guns  

When presenting red flags to pave the road for a conclusion of corruption, counsel and experts must remember that a red flag in itself is not corruption, but an indicator of a potential problem or enhanced risk. Therefore, red flags must always be presented carefully and with consideration of context.

Having set the stage clearly, the expert should then describe the work carried out and set out the pattern of red flags that have been identified. The expert should describe the red flags, detailing how they have been identified and the time period over which they occurred. The pattern of red flags should then be evaluated and summarised. This summary should draw upon the expert’s relevant experience, as well as provide examples of other matters in which similar red flags have arisen.  

Depending on their background and qualification, the expert may not be able to say for sure whether a pattern of red flags affirmatively constitutes a corrupt event (as that can often depend on definitions in relevant statute). However, they should be able to give an opinion as to whether the red flags are consistent with the hallmarks of corrupt transactions.

Experts should keep in mind that the facts will be critically examined and interpreted based on the judgement of the tribunal. Part of reaching the point of a clear presentation of facts and analysis to the tribunal is to select the issues to be considered, even if that sometimes means setting aside parts of the analysis that required significant efforts.

Very rarely will experts find a smoking gun that removes all doubts about the existence of corruption. However, it is up to the experts, and counsel, to present their findings in a manner that highlights the heightened risk of corruption to a point where it is no longer reasonable for the tribunal to ignore it.

Connect with the authors: Derek Patterson, Amina Khoungui, Jorge Lopes and Kim Qvale.

Return to this series’ part 1 tracing the growing salience of corruption evidence in arbitration matters, and part 2 on corruption typologies.

This series is adapted from the authors’ contribution to Global Arbitration Review’s Guide to Investment Treaty Protection and Enforcement, available online.

[2] Sorelec v. Libya, ICC Case No. 19329/MCP/DDA, Paris Court of Appeal, No. RG 18/07347, November 2020

[3] Sorelec v. Libya, ICC Case No. 19329/MCP/DDA, French Court of Cassation, Judgment No. 610 FS-B, September 2022

[4] Valeri Belokon v. Kyrgyz Republic, Court of Cassation (First Civil Chamber), Judgment 17-17.981, 23 March 2022

[5] Lucinda A Low, ‘Methodologies for proving corruption in arbitration: Uses and limitations of red flags’, in Régis Bismuth, Jan Dunin-Wasowicz and Philip M Nichols (eds), The Transnationalization of Anti-Corruption Law (Routledge, 2021).