Michelle Zak, co-founder and managing director at technology provider Qomply, analyses the collapse of US firm Archegos Capital Management, and how its US$10 billion loss in 2021 highlighted the need for the industry to increase the introduction of systemic risk controls
An unfortunate but highly concentrated exposure to a handful of equities ultimately led to the collapse of Archegos Capital Management, a US family office, in March 2021. Approximately US$10 billion in losses were sustained by Archegos’ counterparty banks, including Credit Suisse, Nomura, Morgan Stanley, UBS and Mitsu. This is a reminder that systemic risk still exists as a clear and present danger. Since 1998, with the collapse of Long Term Capital Management (LTCM), there have been a number of initiatives introduced by regulators, aimed at transparency and systemic risk management. Further tightening occurred with the second Markets in Financial Instruments Directive (MiFID II), European Market Infrastructure Regulation (EMIR) and Securities Financing Transactions Regulation regimes which were introduced or fine-tuned to avoid a repeat of the crisis of 2008.
It could be argued that most notable systemic failures have been linked to a significant macro event, with market turbulence acting as a catalyst for correlated contagion. In the case of Archegos, the markets were calmer and quieter, suggesting this disaster was the result of a failure in systems and controls.
However, Archegos slipped below the radar of US regulators as they were outside the reporting requirements of the U.S. Securities and Exchange Commission (SEC) and Financial Stability Oversight Council (FSOC).
In the US, Archegos was classified as a family office and, as such, were not subject to reporting requirements. Archegos was able to put on highly leveraged derivative trades known as total return swaps without having any disclosure requirements.
Family office, by definition, is a reserved, US classification for investment companies that do not deal with the public in an investment advisor capacity, and its dealings are solely for the investment purposes of clients, comprised of family members or family entities.
So how is it that the over-leveraged positions of Archegos were not spotted in the EMIR reports submitted by their European counterparties?
In his paper, Antoine Bouveret, senior economist at the European Securities and Markets Authority (ESMA), openly discusses the usage of EMIR data to highlight the “steep increase in concentrated exposures”, undertaken by Archegos across a two-month period from February to March 2021, which preceded its ultimate demise in March 2021.
Using EMIR data submitted by European counterparties, ESMA identified that five principal stocks accounted for over-leveraging equivalent to 360pct of its net exposure during this two-month period.
According to Bouveret: “Archegos positions with EU counterparties are analysed […] two months before its demise there were warning signs that Archegos had substantially increased its exposures to a few stocks, making the firm highly vulnerable to adverse market developments related to these shares. Our analysis shows how supervisory data can be used for risk monitoring purposes. We also review data gaps and reflect on the regulatory lessons learned from the collapse of Archegos.”
The EMIR transaction reports, submitted by European counterparties, highlight some of the challenges in spotting evidence of systemic risk using valuations and margin requirements in advance of a widespread unwinding. When reviewing data retrospectively, it is easier to spot the issue. However, analysing massive datasets looking for correlated risk exposure in real-time is a resource-intensive undertaking.
ESMA has not specifically provided comments regarding Archegos and whether their monitoring algorithms had flagged any breach in exposure thresholds. It may be safe to presume that the event may have gone undetected.
For the efficient surveillance of systemic risk to occur, there is a reliance on data quality and systems and controls – from the counterparty level to the local jurisdictional level, and to an inter-jurisdictional level in the case of ESMA and the UK Financial Conduct Authority (FCA).
According to Zach Johnson, director of regulatory consulting at Kroll: “This is precisely why the regulators harp on about ‘data quality’, and the need for firms to have the appropriate systems and controls in place.”
“By the time firms engage us, 50 per cent of them have already received a tap on the shoulder by the regulator with regards to poor data quality and the need to tighten their systems and controls. Being on the front foot would not only avoid any potential regulatory scrutiny, but also assist in the wider agenda of effective monitoring and surveillance, so that the entire industry benefits,” continues Johnson.
To this end, there has been a continued focus on data quality and systems and controls as the underlying themes. ESMA’s work programme for 2022 certainly serves as evidence as it ranks data quality highly across various initiatives.
In the work programme, ESMA outlines plans to develop “supervisory methodologies for data analysis and quality monitoring”, as well as the “enhancement of data analytical systems and methodologies including preparing future use of machine learning techniques and big data infrastructure, and the possible delegation of data quality activities.”
The work programme specifically notes that “ESMA will continue monitoring the consistent implementation of EMIR in relation to over-the-counter derivative requirements, in particular regarding clearing and bilateral margin requirements.”
In April 2022, ESMA released a data quality report that highlighted some of the findings in their review of EMIR reporting. Entities exceeding an error tolerance level of 1pct on aspects of their reporting were notified by ESMA.
According to the Authority, 15 counterparties across seven jurisdictions were notified for follow-up. They have published their findings of jurisdictions that have the highest number of abnormal values in EMIR, reporting across key valuation and margin metrics.
Malta, Germany, France, Denmark and Cyprus ranked amongst the highest with abnormal values, although it is unclear as to how metrics were applied; that is, by either percentage of reports, or by absolute numbers. Either way, Malta and Cyprus appearing in the list against larger jurisdictions certainly raises an eyebrow.
In the UK, the FCA has not yet released any further statements regarding the data quality specifically relating to EMIR reporting. Although, if ESMA and the FCA initiatives are in-step, then perhaps it is simply a matter of time.
It is clear that EMIR transparency reports provide regulators with an artillery of data. Threading together valuation and margin data for underlying instruments across entities can be effective in identifying areas of concentrated exposure.
Johnson concludes: “The challenge for regulators now remains in calibrating surveillance tools and training algorithms to identify issues before they become unmanageable. As for the firms, whilst they may not have an overarching view of exposure, they can certainly be more proactive in identifying issues by implementing the appropriate level of systems and controls, while taking quick corrective action when thresholds are breached.”
