Overly generous risk weightings on certain transactions, a lack of data and stress testing loopholes means that some global systemically important banks could get hit by another Archegos-type failure. By Justin Pugsley 

The fall of the family office Archegos Capital Management in March 2021 was a classic tale of hubris, excess leverage and deliberate deception. The firm, which sat outside much of the regulatory perimeter, used total return swaps based on US listed media and tech stocks to juice leverage and hide its positions from the market. 

Meanwhile, the saga revealed ongoing weaknesses in the regulatory framework, meaning that banks must remain on top of their risk profiles.

Credit Suisse and Nomura Securities were the hardest hit, but others such as UBS, Mitsubishi UFJ and Morgan Stanley were also impacted, notching up total losses of more than $10bn. It is one of the biggest financial trading losses in history.  

To avoid a repeat: “I would advise bank boards to make sure a sound stress testing methodology is in place, which works with sufficiently large instantaneous shocks.” says Andreas Ita, managing partner at Orbit36, a financial risk consultancy.

He also advises boards to set sound capital allocation frameworks that properly price risk. “Archegos has shown that the margins some banks earned are lower than their cost of capital,” he says. For most global systemically important banks (G-SIBs), prime brokerage is a low-margin business, often treated as an add-on service for big clients. 

SA-CCR gaps

Despite greater vigilance by supervisors and bank risk officers, Mr Ita warns that banks could still suffer big trading losses. “I think [G-SIBs] remain vulnerable because margins and risk weighted assets, particularly for collateralised over-the-counter (OTC) derivatives are relatively low, since they are, in our view, not calibrated conservatively enough,” he says.

Mr Ita, a former G-SIB capital optimisation head and equity derivatives trader, believes there are some gaps in the standardised approach for measuring counterparty credit risk (SA-CCR). 

Though he thinks the revised methodology is basically sound, he points out that for equity swaps and equity derivatives in general, a standardised decline in prices is implicitly assumed. 

But with Archegos, there were bigger-than-anticipated falls in underlying stock prices, seeing banks rack up big losses as the family office defaulted on its margin commitments. The situation was made worse as banks tried to liquidate positions in the same underlying stocks simultaneously. 

“The problem is that [for SA-CCR] it doesn’t matter whether it’s a highly liquid blue chip stock an illiquid stock. It doesn’t matter how much of the daily volume you hold on your books,” he says. He argues that there should be a capital add-on to reflect higher risks associated with illiquid or concentrated positions. He thinks it will be years before the SA-CCR is refined to reflect these risks because the Basel Committee on Banking Supervision only recently finalised the framework. 

Mr Ita will delve deeper into the ongoing risks and solutions for G-SIBs related to the Archegos saga in a workshop at the Capital Management for Banking Institutions conference on November 30-December 2 in London.

Stress test arbitrage 

Another reason why Archegos managed to slip through the net was due to differing stress testing regimes in some jurisdictions.

G-SIBs tend to use central booking hubs for their trading activities — a legitimate practice that allows them to efficiently manage their risk. London is favoured for housing OTC derivatives trades. 

Mr Ita believes that the US Comprehensive Capital Analysis and Review (CCAR) stress tests would have revealed the Archegos exposures and therefore would have triggered a higher capital charge in the US. 

However, foreign banks only have their US operations stress tested under CCAR, while for US banks it is the consolidated balance sheet. According to Mr Ita, the EU stress tests would have detected this as well. But since the UK left the EU, it has its own stress testing regime. 

“In the UK, you have completely different rules for margined trades,” he says, adding that the UK changed its stress test guidance to exclude stressing collateralised equity derivatives positions in 2021. 

UK regulators were satisfied that daily margining would cover potential derivatives losses. But the Archegos failure showed that margins can offer insufficient protection under some circumstances.  

“That means the same exposure, which would have triggered a very high capital charge in the US under CCAR, would attract a zero-stress capital charge in the UK,” he says. “We don’t think it is a coincidence that this mainly hit Swiss and Japanese banks.” Surprisingly, UK banks largely escaped, perhaps a legacy of being subject to EU rules until December 31 2020. 

Mr Ita says the foreign banks did their booking mostly in London, outside the CCAR’s and EU stress testing coverage, and the UK stress-testing methodology could be a reason why they did not sufficiently capitalise their US Archegos-related positions.  

“It will be interesting to see if the Prudential Regulation Authority changes its guidance in the upcoming UK stress tests,” he says. 

In May, the European Securities and Markets Authority released a report calling for more data gathering from national competent authorities to detect potential information gaps. “I’m a little bit sceptical, of course it helps,” says Mr Ita, adding that much depends on the speed and ability of regulators to detect problems. “But you have to remember that Archegos built up their positions in a way to intentionally conceal their exposures.” The entrails of the Archegos scandal will be picked over for years in a bid to avoid a repeat.

The Financial Times would like to sponsor your first month’s subscription to Global Risk Regulator. To start your no-obligation trial, please contact: