Collar financing, low op risk losses and ESG data

Asia collar financing surges on back of Covid-19 volatility

Options-based structures gain ground on margin loans – and dealers say it may be a structural shift

Op risk data: losses plummet during lockdown

Also: Wirecard dominates legacy losses; SEB hit with massive AML fine. Data by ORX News

Investors turn to raw data over ratings in ESG alpha hunt

Quants are using data on product returns and employee welfare to pick winners

COMMENTARY: Too quiet

The human disaster of 2020 has been mirrored in the macroeconomic data – the lists, lengthening each day, of the sick and dead of the pandemic are echoed by reports of rising unemployment, economic contraction, business failure and spiking debt.

In that context, one report on Risk.net this week might have come as a surprise – total operational risk losses for the first half of 2020 were only just over half the figure for the same period in 2019, just $7.39 billion against $14.28 billion.

If by this point in 2020 you’re instinctively suspicious of anything that looks like good news, well, you’re thinking along the right lines. This does not in fact represent an unexpected and welcome improvement in the ability of financial institutions to stop tripping over their own feet. It doesn’t even reflect that business has slowed with the economy – fewer transactions, fewer loans, less activity generally – and so there are fewer operational risk loss events because there are simply fewer operational events (in the way that road traffic deaths dropped after the lockdown because people were simply travelling less).

In fact, as loss data provider ORX points out, “evidence suggests regulators around the world have focused more on helping firms weather the crisis instead of imposing fines and penalties”. There’s also been a shortage of timely reporting – banks trying to stay afloat and manage a largely remote workforce have higher operational priorities than sending their loss event details to ORX when they ought to.

And it’s also possible that this has affected their ability to detect loss events in the first place. Data from the last great recession in 2008–9 showed an average lag of 30 months between the initiation of a fraud loss event and the discovery of the fraud – the loss that shows up in 2019 might well be the result of a control failure in 2016 or earlier. And there are good reasons to expect an increase in fraud in the first half of 2020 – disrupted by the lockdown, fraud prevention offices will be operating less efficiently, and their systems will be plagued by false alarms caused by (non-fraudulent) shifts in the behaviour of their customers and colleagues.

Even as the pandemic continues, financial institutions are looking at long-term changes in the way they work, from increased remote working to a reduction in outsourcing, and preparing to handle the operational challenges these changes will bring. They need to be alert to fraud as they do so – the 2020 spike in operational risk losses has been hidden, not avoided, and will show up in the loss data sets and in the bottom line for many months and years to come.

STAT OF THE WEEK

Under a doomsday scenario set in March 2019 in which the CCP’s top two clearing members collapsed, LME Clear would incur about €712 million ($808 million) of losses beyond the required initial margin of the defaulters, exceeding its €446 million of prefunded guaranty resources. The additional losses would have to be mopped up using further contributions from healthy clearing members.

QUOTE OF THE WEEK

“The [ABN Amro] corporate and investment banking business consumes a third of group capital. It’s around 20% of group revenues, but it’s 70% of the risk cost over the past five to seven years. They have this huge volatility and risk cost – be it clearing and commodity finance in Q1, be it diamonds in the last two years, or be it shipping – that causes a disproportional amount of credit losses” – Johan Ekblom, UBS