![Risk.net](https://www.risk.net/sites/default/files/styles/print_logo/public/2018-09/print-logo.png?itok=1TpHrpuP)
This article was paid for by a contributing third party.More Information.
Collateral damage: the lowdown on dirty CSAs
![Collateral damage: the lowdown on dirty CSAs](/sites/default/files/styles/landscape_750_463/public/2024-01/GettyImages-487730228.jpeg.webp?h=91d5c1b1&itok=hUUglaus)
How are banks coping with growing demand for non-cash collateral in uncleared derivatives contracts? An expert panel discusses the re-emergence of dirty credit support annexes (CSAs)
The panel
- Allan Cowan, Head of financial engineering for financial risk analytics, S&P Global Market Intelligence
- Nicki Rasmussen, Head of XVA desk, Danske Bank Corporates and Institutions
- Magnus Lindahl, XVA specialist, Nomura
The pricing and risk management of uncleared derivatives are rarely straightforward, but recent calls for a wider range of collateral to be used in these agreements – so-called ‘dirty CSAs’ – are muddying the waters.
Banks favour CSAs in which counterparties only post cash to each other – making it easier for incoming collateral to be reused for outgoings. But 2022’s meltdown in the gilt market, following the disastrous UK mini-budget announced by then-prime minister Liz Truss, left many pension funds scrambling for greater asset flexibility. Dirty CSAs allow clients to post corporate bonds as well as government debt and cash – and demand is growing, following a trend that began with some insurers needing to post bonds as CSA collateral following the Covid-19 pandemic.
Such agreements make it harder for banks to predict their own funding needs and associated risks. In addition, capital, leverage and funding regulations impose direct costs on bank holdings of corporate debt. This adversely affects derivatives exposures under Basel III’s leverage ratio, risk-weighted capital under the standardised approach to counterparty credit risk and funding costs under the net stable funding ratio.
The cost of these various charges is typically applied to the upfront cost of each trade, making it more expensive to use a dirty CSA.
While banks have been generally more accepting of a wider range of assets, these factors mean the impact of poor decisions can be amplified, with a substantial impact on profitability both at the time of the trade and further down the line.
Demand has risen for dirty CSAs – should banks be more accepting of a wider range of collateral used in these agreements?
![Allan Cowan, S&P Global Marketing Intelligence](/sites/default/files/styles/medium/public/2023-11/Allan%20Cowan%2C%20SandP%20GMI%202023.jpg.webp?h=3e7b9896&itok=L0Z-MflD)
Allan Cowan, S&P Global: Simpler CSAs that banks have been pushing towards over the past decade are better for them so they will need to think carefully before going down the dirty CSA route, as it adds various risks and challenges. Demand is being driven by their clients though, so banks are likely to have to accept some of this business to stay competitive.
It is not restricted to a particular region so, in the longer term, more of this business may favour larger banks with a global footprint and more developed systems that are more comfortable managing additional risks and pricing options appropriately.
Magnus Lindahl, Nomura: We’ve seen a big push for this type of agreement from clients, and there will always be a market for them. Being able to post those bonds was clearly very helpful for pension funds. They are now significantly more expensive with no netting between the collateral and the mark-to-markets, so they need to be priced appropriately.
Nicki Rasmussen, Danske Bank Corporates and Institutions: The return to dirty CSAs reverses a trend I've seen for many years of trying to simplify CSAs because end-user clients wanted transparency and less complexity.
But that simplification has come at the cost of flexibility, and has resulted in the reduced ability to manage liquidity. Under dirty CSAs, clients have to live with pricing implications and the risk of being in a difficult position when they want to unwind trades – something that has been seen previously.
Danske Bank was very much in favour of the simplification trend so it would take a lot to convince us to change back. But we want to service clients, so would always discuss the trade-off and assets available, whether they have access to repo, and so on.
Pension fund clients in the Nordics mostly look for simpler CSAs incorporating high-grade government bonds. That is still manageable because those are fairly liquid assets that can be sourced and reused in most market conditions.
What issues do dirty CSAs raise for banks?
Allan Cowan: The main challenge is pricing. Banks sell an option to the client to post whatever collateral they want, so pricing that risk and controlling it is probably the biggest challenge.
Many banks have controls around how much of a particular asset can be posted, which become more restrictive as the amount of collateral rises. They will also favour types of collateral they can turn over quickly or that are easiest to rehypothecate.
Another issue is the potential for wrong-way risk between collateral, which is dependent on market value and exposures. This can create additional risk to valuation adjustments – known collectively as XVAs – calculations that need to be modelled appropriately.
![Nicki Rasmussen, Danske Bank Corporates and Institutions](/sites/default/files/styles/medium/public/2024-01/Nicki%20Rasmussen%2C%20%2CDanske%20Bank.jpg.webp?h=cbeadb1c&itok=mxJ7t_no)
Nicki Rasmussen: A CSA is, in some sense, a committed repo facility where you pledge to do an almost unlimited notional - at least a notional that is proportionate to the risk in the underlying portfolio - and with the duration equal to the longest trade in the netting set you are collateralising.
By committing to accept whatever collateral is in the CSA, you cannot renegotiate because there is no price discussion other than when you add new trades or terminate trades.
So, if you have a large derivatives portfolio with trades running for 30 years and can post corporate bonds – or whatever is in the CSA – then the bank has given you a strong commitment that you could not buy as a separate product.
You could not call up a repo desk and say: ‘I'd like a 30-year commitment to always post you up to the mark-to-market figure of this swap’ – no bank would commit repo lines to something like that. But most banks have no issue signing a CSA that implicitly has the same commitment.
If a bank accepts these deals with a particular group of clients but doesn't have any offsetting trades, the assets can end up on the balance sheet and need to be repo'd out or find another solution. If all of those clients have undertaken receiver swaps, you want a segment of clients that have done payer swaps with the same CSA so you can recycle it.
In the best of worlds, you want a balanced book but, with dirty CSAs, the bank is left to deal with them in its liquidity risk management.
How can banks best optimise collateral in the current environment? What new tools and techniques are available?
![Magnus Lindahl - Nomura](/sites/default/files/styles/medium/public/2018-04/Magnus%20Lindahl%20-%20Nomura.jpg.webp?h=9be45261&itok=ctkFpFyl)
Magnus Lindahl: Focusing on risk-reducing trades and optimisation is very much business as usual for Nomura. We spend a lot of time reviewing, setting up and sourcing different types of collateral that can be used from different parts of the organisation. We make sure we have the right collateral embedded within the different CSAs and try to get more specific products into some to make it more efficient.
Nicki Rasmussen: What you need to know for any given collateral is: what are the risk factors driving the collateral balance? I can look at data showing types of collateral in one direction and underlying risk factors in the other. I can see whether, for example, euro rates would drive demand for dollar cash collateral up or down and among which agreements. Do they net out or should I try to rebalance or do more of certain business? It's super-complex and not something you can optimise down to the penny.
You can try to control the risk by setting limits on the use of flexible collateral in the contract. You can also engage in trades with set CSA terms that, under certain market conditions, will give you collateral of a particular type, and then do offsetting trades that demand you return collateral of another type. These liquidity switch trades have been seen since before the financial crisis [that began in 2007–08]. The tricky bit about those trades is that most banks would like to be in the same position, so executing them among the banks is difficult.
During the financial crisis, some clever banks approached treasury departments or pension funds, offering them an attractive few basis points to do those trades. But they didn't fully realise the value and offered that liquidity transformation too cheaply, especially when markets became stressed and the good collateral became very expensive and the bad collateral became very cheap. So there is a warning there in being too attracted by a few basis points to do such a trade.
Allan Cowan: As pricing is the toughest challenge, you need XVA systems to support you in calculating a cheapest-to-deliver spread for collateral valuation adjustments (ColVA) that properly prices this option.
It’s not just a question of which is the cheapest asset – because of the controls banks have in place and how much of a particular asset they're willing to take. Constrained optimisation of the cheapest-to-deliver spread is a key tool and one that is very much in demand to help clients calculate ColVA under dirty CSA situations.
You also need to be able to model the assets you've taken as collateral. Haircuts are one, more conservative, way to approach this and can do well in terms of over-collateralising the bank concerning the risk in those assets.
But what this often misses is strong wrong-way risk effects where the collateral may jump at default – particularly if it’s a large corporation strongly reliant on the health of the national economy, such as in emerging markets. You can find yourself quite under-collateralised if markets move against you.
Get instant XVA measures with zero implementation headaches
The XVA Pricer by S&P Global Market Intelligence provides deal-time XVA measures powered by a best-in-class analytics engine and market data. No data integration required.
Now you can instantly access the full power of the S&P Global Market Intelligence XVA solution with just a simple login.
Sponsored content
Copyright Infopro Digital Limited. All rights reserved.
As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (point 2.4), printing is limited to a single copy.
If you would like to purchase additional rights please email info@risk.net
Copyright Infopro Digital Limited. All rights reserved.
You may share this content using our article tools. As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (clause 2.4), an Authorised User may only make one copy of the materials for their own personal use. You must also comply with the restrictions in clause 2.5.
If you would like to purchase additional rights please email info@risk.net