Counterparty Risk – A Reality Check

Recent events have provided just such a check, with the failure of Silicon Valley Bank (SVB) in the US, and Credit Suisse. This call was a “lessons learned” session, with an excellent and very detailed discussion on the operational aspects of what people went through, together with some of the broader issues, including the, very human, tendency not to take risks seriously until they materialise.

The call took place before the failure of First Republic Bank, and its acquisition by JPMorgan.

This summary is two pages long – this was a very detailed call, with a lot of practical learning and experience sharing.

So I will put the bottom line before the long read: the participants who had problems with the collapse of SVB did so because they had delayed integrating recent acquisitions into their normal counterparty risk management processes. With acquisitions, there are good reasons for doing this – but also risks. Others felt management was losing sight of these issues – so it was a useful wake-up call. All participants were grateful to their centralised processes, which meant they could get the information on the exposure quickly, even if this required working with other, centralised, functions. All participants are moving away from using smaller regional or local banks, wherever possible.

It can be hard to justify having a proper and rigorous counterparty risk management process, especially when it has been a few years since the last significant bank failure. This was an important wake-up call. Fortunately, it came at no cost.

Operational issues:

• Generally, SVB created more anxiety, as the failure became apparent on a Friday, and the intervention of the US federal authorities only became certain over the weekend.
• Everybody scrambled to understand what payments were going into SVB, so they could be put on hold. The concern was that, if you pay employees or suppliers by making a transfer into an account with a failed bank, they may not be able to access the cash. This concern was removed within a couple of days, so payments resumed.
• One participant’s company asked customers to put payments into an account with SVB on hold, and only make payments if there was an alternative bank available.
• No participant reported having a good process set up to get the information about payments going into SVB – this information typically resides in HR for payroll and Accounts Payable for supplier payments. However, after mobilising all the relevant teams, and with help from IT, most participants were able to get the relevant information by the end of Monday. The existence of shared service centres greatly eased this process.
• A couple of participants had an unexpected problem: SVB was not a key relationship bank, and, by policy, they did not have accounts with them. However, these policies had not yet been implemented in recently acquired companies, which did have accounts there. Further, these recent acquisitions were not yet included in the group cash reporting structure, so the information was not readily available as to what the exposure was.
• Another participant had a recently acquired subsidiary which had not yet been brought into the global cash pooling and sweeping mechanism. Their cash was with SVB….
• One participant reported an exposure via SVB’s relationship with HSBC in the UK, and not being aware of it until the news broke.
• All participants rushed to transfer all their cash from SVB to other banks, even though this helped to ensure SVB’s failure. The psychology of bank runs has not changed.
• With Credit Suisse, one participant is trying to move operations elsewhere – but is not sure this will actually be possible, given how deeply rooted and pervasive the relationship is.
• The same participant also has a loan from Credit Suisse, which contains a right of offset. While there was no need to exercise this right, it was very valuable, and reassuring, to have this unusual provision in the contract.
• Generally, there was less panic around the Credit Suisse situation, in part, because the UBS takeover was clearly signalled very quickly; in part, because there had been warning signs for several years, so most participants had diversified their risk.

Counterparty Risk Management:

• Do not assume things are safe without challenging the assumptions. Several participants had viewed the US bank system as relatively safe, and so were less strict on their counterparty risk management there.
• There can be delays in integrating acquired entities into the counterparty risk and bank management system. This can result in being exposed to this kind of event, as happened to more than one participant.
• Generally, these events have confirmed that most participants will limit their exposure to their main relationship banks, i.e., the larger international institutions. This is unlikely to be beneficial to smaller, regional banks.
• Linked to this is the question of what happens if one of the larger banks gets into trouble. One participant is already uncomfortable with their exposure to UBS, which is now larger than Switzerland’s GDP – and Iceland demonstrated a few years ago that a country can be unable to bail out its banks. Still, the feeling was that there are banks who are simply too big to fail – and they will be the ones used going forward.
• How do you assess the counterparty limit for a bank?
  • Following their failures which contributed to the Global Financial Crisis in 2008, the rating agencies again demonstrated that a good rating does not prevent a bank run and failure. Treasurers recognise this, but usually lack the resources to do their own, independent, assessment.
  • Should you get into doing your own assessment of how a bank manages its operations, over and above the usual financial indicators? One participant thinks they should do this – but all have issues with the resources required to do so.
  • One participant determines counterparty limits relative to a bank’s capital. If this is too large a figure, they do it in relation to their company’s own equity.
• There are markets where companies are required to work with local banks, who would not normally have a limit assigned to them. The policy here is to keep the exposure as low as possible.
• The view was expressed that this has helped make sure that counterparty limits and risk management are taken seriously – there was a feeling that, since the peak interest in 2008, people may have started paying less attention to these risks
• One participant is making sure that they run the numbers daily and are warned every time the exposure with a bank approaches the counterparty limit.