The path to optimisation
22 August 2018
John Southgate of Northern Trust suggests the changing regulatory landscape has caused an increased demand on client collateral
Image: Shutterstock
What were Northern Trust’s motivations behind the new collateral optimisation solution? And how does it work?
Initially, we are only offering collateral optimisation to active collateral management clients, which makes sense because effectively the solution enables us to make more intelligent decisions about which assets to select when covering margin calls on their behalf. The changing regulatory landscape has caused an increase in the demands on the client’s collateral.
They are having to give up greater levels of their assets, typically of the high-quality liquid variety, and in some cases, they need to raise cash where they have cash only CSA’s. If they do not have sufficient cash they may have to go into the financing markets to use repos, for example, to access cash against the assets that they have got; that itself is a margined product.
All paths are leading to the fact that margining is generally getting more onerous and complex, therefore, a client needs to increase the efficiency with which they use their assets. We had some clients going back two or three years ago that were identifying this complexity. They provided us rules over which assets to select but it wasn’t algorithmic and not as flexible as an automated solution. We also saw some consultants up to three years ago asking “what is your collateral optimisation strategy?”. However, I’m not sure whether all of the consultants out there knew necessarily what they were asking because optimisation means different things to different people—but we knew we needed to have a defined strategy.
In terms of how we define collateral optimisation, in the first instance, we have established a holistic view across client’s assets and across their margined product silos, which we call the enterprise inventory view. Effectively, that ensures that we’ve got accurate data in terms of availability of the client’s assets and their margin requirements across each of those silos, for example over-the-counter (OTC), foreign exchange (FX), repos, exchange derivatives, and so on. Only once we have that view, may the optimisation algorithms be applied. Once the enterprise inventory view was in place we developed a few flavours of what we call cheapest-to-deliver algorithms. This takes in factors such as haircuts on underlying assets, maturity and funding costs, to try and identify, for a given grouping of margin requirements, what the cheapest or most efficient assets are to deliver. It is important to note that we don’t look at each individual margin call separately but collectively, depending on the logical grouping a client wants to use.
For example, where we are the collateral manager for a large investment manager, they will have lots of underlying funds and each of those funds are facing off to multiple counterparties. Given that it’s only the fund’s assets that we can use, we will optimise daily margin calls at that level. Effectively, the algorithm attempts to find the right assets for the most difficult agreements, or the hardest to fulfil first, rather than just going sequentially one-by-one. The ability to re-optimise entire collateral portfolios is also useful for some clients. This looks at everything that has previously been delivered across all of their counterparties and then pumps it through one of the algorithms and says, ‘if you have an unlimited restriction around the number of substitutions that you wanted to make this represent the most efficient set of assets to deliver’, then there is an exercise where we would recall those assets and substitute. Typically, clients would be doing that on a less frequent basis, maybe monthly or quarterly. That’s part of the optimisation offering as well.
We are seeing some pretty compelling numbers in terms of savings that can be delivered. While we have built an initial set of standard algorithms, every client is different, which is why we are able to overlay client hierarchy rules which tell us the sequence of assets to use and allow for customisation. Those rules can be broad brush or very granular and are fully automated. If there are specific assets that the client doesn’t want us to use, for example, we can strip them out of scope from the algorithms.
What benefits/opportunities can the collateral optimisation solution bring to clients and prospective clients?
It is becoming a core offering of our active collateral management service. When we talk to clients this is often at the forefront of discussions, however, it is worth calling out that optimisation isn’t critical for every type of client. If you have only got a very small number of counterparties there is probably very limited scope for optimising your portfolio and if clients have got cash-only CSA’s then they can’t be optimised in the same manner.
One of the key benefits that we are seeing with live clients is that it is reducing the overall market value of assets that we are delivering out to cover margin calls, which means that the client has more assets left which they can leverage to enter into liquidity trades to raise cash if they need it, or more assets through which they can lend, so there is revenue opportunity there for them. Some clients may hold back from securities lending because they want to ensure that they have a sufficient buffer of assets to cover their margin calls. With optimisation, because you are looking more holistically across those silos, you can be a lot more joined up so you can be more confident that you are going to have sufficient assets left to meet the margin requirements. And because you are delivering reduced market value out to counterparties you are reducing your credit to counterparty exposure, there are less assets sitting at the counterparty.
Who is it best targeted at?
It is most effective when you have a number of counterparties, say five or six, and multiple underlying products that you are margining. So effectively, the more complex the client’s collateral book is, the more potential benefit there is through optimisation.
Why are some firms still hovering around the issue of collateral optimisation? And maybe not utilising it to its full potential?
This has been a significant investment for us and we have been working on implementation throughout last year. We are now working on the next phases and have got multiple initiatives in terms of increasing the functionality. That investment may be holding some people back and it is not suited to all types of firms so the cost either in-house or via a third party might not warrant the benefit that you are going to get. Therefore, when we work with prospective clients we look to run through a parallel period with them, where we can show them the benefits that we get through optimisation. They can make a call in terms of what the value is to them, whether the market value reduction is sufficient, and whether that benefits other investment activities. We are very happy to help them go through that discovery process.
What are you hearing from clients in the collateral space? And what matters most when optimising collateral and controlling liquidity?
It’s certainly that holistic view that I’ve spoken about, where you need to break down the silos. It’s no good managing pools of collateral for different types of asset classes in isolation. And actually, we have seen over the past couple of years that there is increasingly more desire to centralise collateral management. Historically, a large asset owner who might have multiple investment managers managing assets for different portfolios may have let each of those manage the collateral on their behalf. That is simply not efficient because you’ve got the same funds effectively facing off with the same counterparties and you have multiple pools of collateral.
There are no netting benefits. Without centralisation, it is difficult for an optimisation strategy to be effective. When optimising, reliable and timely data is absolutely necessary. It is important to ensure that we have got the latest snapshot of what assets are available, without time lags, especially if assets are being actively traded.
Additionally, we need to make sure that we are not stepping on the front office’s toes. Some of the things that we have developed have allowed us to ensure that if an asset manager knows that they are going to be trading a set of assets over a period of, say a week, that we can automatically remove them from optimisation selection, and take them off the table for a period of time. Clients are starting to think about collateral much more from a front office perspective, and it has certainly been spoken about for a number of years—collateral management moving out of the back office towards the front office. Decisions are starting to be made around “what’s the impact to my margin requirements by doing this particular trade or clearing via this central counterparty?” so, it is about looking at the bigger picture and the impacts to performance that collateral can have.
Have regulatory changes challenged the way firms think about collateral management?
The increasing regulatory requirements have added complexity and that is certainly bringing collateral management to the forefront of people’s minds. The decisions about where you might clear a trade can make an impact on the netting benefits at a clearinghouse. Consequently, we are starting to see more scenario analysis tools available, certainly, that’s where we want to get with this tool.
Right now it sits within our organisation, but in future phases, we are not ruling out being able to put it on clients’ desktops so that they can actually run scenario analysis across their book. We are not there yet but that is an aspiration for sure. Some of the clearing members of OTC clearing house offer tools, it is essentially risk scenario analysis but looking at margin implications. The un-cleared margin rules last year were a big impact and transformational for a lot of clients that hadn’t had to previously exchange margin but suddenly found themselves needing to.
We have spoken to a lot of firms that never really had to consider the fact that they might need to hold cash or collateral back for margin. It’s about understanding the impact of a certain trade on your collateral requirements, and then what is the cost of that, particularly if you have got to access cash financing markets and how does that cost impact overall performance.
Have you had positive feedback on the solutions from clients?
We are seeing really encouraging results in the reduction of market value that we are delivering. We have revisited some clients margin portfolios once they have been live for a few months and we have seen some positive numbers. Client feedback has been very positive and those who have started on the path of rolling out want to get the rest of their funds optimised as soon as possible, so we are now working through a pipeline of clients to get them on to the tool.
How do you see the collateral landscape changing over the next five years?
The theme of the increased complexity is not going to go away as there are further regulatory challenges ahead. The new initial margin rules are going to kick in over the next few years and will impact our clients, so I think that there is a lot going on in the industry right now where we are looking to increase messaging automation, and that is the other aspect of efficiency.
I look into efficiency in two key areas. One is an efficient use of the assets that you’ve got, which we call optimisation and the other is infrastructure and messaging, which is the efficiency and the speed at which we can move assets around to cover margin calls. There is a lot of exciting stuff going on around that. I hope that the industry will continue to adopt some of these because standardisation and streamlining is clearly a good thing.
The theme around liquidity is going to continue to grow. We have got clients that sit on both sides of the fence in terms of some having a potential cash shortfall, whereas there are others that are long cash. Interesting models like peer-to-peer lending are becoming more important and we are starting to see more opportunities emerging.
That will continue and, of course, with these trades requiring margining, that trend of increasing complexity will increase. I have spoken about centralisation, and I think that will clearly continue. Modelling from a front-office perspective will become more prevalent as will the tools used to help model the different scenarios around what a specific trade does and what that means from a marginal cost perspective, what that does to your profit and loss, and almost having a front office role in terms of the collateral management function alongside investment management. We certainly see liquidity management and collateral management being two areas that sit pretty neatly together, and that is how we think about some of the products that we look to offer.
Initially, we are only offering collateral optimisation to active collateral management clients, which makes sense because effectively the solution enables us to make more intelligent decisions about which assets to select when covering margin calls on their behalf. The changing regulatory landscape has caused an increase in the demands on the client’s collateral.
They are having to give up greater levels of their assets, typically of the high-quality liquid variety, and in some cases, they need to raise cash where they have cash only CSA’s. If they do not have sufficient cash they may have to go into the financing markets to use repos, for example, to access cash against the assets that they have got; that itself is a margined product.
All paths are leading to the fact that margining is generally getting more onerous and complex, therefore, a client needs to increase the efficiency with which they use their assets. We had some clients going back two or three years ago that were identifying this complexity. They provided us rules over which assets to select but it wasn’t algorithmic and not as flexible as an automated solution. We also saw some consultants up to three years ago asking “what is your collateral optimisation strategy?”. However, I’m not sure whether all of the consultants out there knew necessarily what they were asking because optimisation means different things to different people—but we knew we needed to have a defined strategy.
In terms of how we define collateral optimisation, in the first instance, we have established a holistic view across client’s assets and across their margined product silos, which we call the enterprise inventory view. Effectively, that ensures that we’ve got accurate data in terms of availability of the client’s assets and their margin requirements across each of those silos, for example over-the-counter (OTC), foreign exchange (FX), repos, exchange derivatives, and so on. Only once we have that view, may the optimisation algorithms be applied. Once the enterprise inventory view was in place we developed a few flavours of what we call cheapest-to-deliver algorithms. This takes in factors such as haircuts on underlying assets, maturity and funding costs, to try and identify, for a given grouping of margin requirements, what the cheapest or most efficient assets are to deliver. It is important to note that we don’t look at each individual margin call separately but collectively, depending on the logical grouping a client wants to use.
For example, where we are the collateral manager for a large investment manager, they will have lots of underlying funds and each of those funds are facing off to multiple counterparties. Given that it’s only the fund’s assets that we can use, we will optimise daily margin calls at that level. Effectively, the algorithm attempts to find the right assets for the most difficult agreements, or the hardest to fulfil first, rather than just going sequentially one-by-one. The ability to re-optimise entire collateral portfolios is also useful for some clients. This looks at everything that has previously been delivered across all of their counterparties and then pumps it through one of the algorithms and says, ‘if you have an unlimited restriction around the number of substitutions that you wanted to make this represent the most efficient set of assets to deliver’, then there is an exercise where we would recall those assets and substitute. Typically, clients would be doing that on a less frequent basis, maybe monthly or quarterly. That’s part of the optimisation offering as well.
We are seeing some pretty compelling numbers in terms of savings that can be delivered. While we have built an initial set of standard algorithms, every client is different, which is why we are able to overlay client hierarchy rules which tell us the sequence of assets to use and allow for customisation. Those rules can be broad brush or very granular and are fully automated. If there are specific assets that the client doesn’t want us to use, for example, we can strip them out of scope from the algorithms.
What benefits/opportunities can the collateral optimisation solution bring to clients and prospective clients?
It is becoming a core offering of our active collateral management service. When we talk to clients this is often at the forefront of discussions, however, it is worth calling out that optimisation isn’t critical for every type of client. If you have only got a very small number of counterparties there is probably very limited scope for optimising your portfolio and if clients have got cash-only CSA’s then they can’t be optimised in the same manner.
One of the key benefits that we are seeing with live clients is that it is reducing the overall market value of assets that we are delivering out to cover margin calls, which means that the client has more assets left which they can leverage to enter into liquidity trades to raise cash if they need it, or more assets through which they can lend, so there is revenue opportunity there for them. Some clients may hold back from securities lending because they want to ensure that they have a sufficient buffer of assets to cover their margin calls. With optimisation, because you are looking more holistically across those silos, you can be a lot more joined up so you can be more confident that you are going to have sufficient assets left to meet the margin requirements. And because you are delivering reduced market value out to counterparties you are reducing your credit to counterparty exposure, there are less assets sitting at the counterparty.
Who is it best targeted at?
It is most effective when you have a number of counterparties, say five or six, and multiple underlying products that you are margining. So effectively, the more complex the client’s collateral book is, the more potential benefit there is through optimisation.
Why are some firms still hovering around the issue of collateral optimisation? And maybe not utilising it to its full potential?
This has been a significant investment for us and we have been working on implementation throughout last year. We are now working on the next phases and have got multiple initiatives in terms of increasing the functionality. That investment may be holding some people back and it is not suited to all types of firms so the cost either in-house or via a third party might not warrant the benefit that you are going to get. Therefore, when we work with prospective clients we look to run through a parallel period with them, where we can show them the benefits that we get through optimisation. They can make a call in terms of what the value is to them, whether the market value reduction is sufficient, and whether that benefits other investment activities. We are very happy to help them go through that discovery process.
What are you hearing from clients in the collateral space? And what matters most when optimising collateral and controlling liquidity?
It’s certainly that holistic view that I’ve spoken about, where you need to break down the silos. It’s no good managing pools of collateral for different types of asset classes in isolation. And actually, we have seen over the past couple of years that there is increasingly more desire to centralise collateral management. Historically, a large asset owner who might have multiple investment managers managing assets for different portfolios may have let each of those manage the collateral on their behalf. That is simply not efficient because you’ve got the same funds effectively facing off with the same counterparties and you have multiple pools of collateral.
There are no netting benefits. Without centralisation, it is difficult for an optimisation strategy to be effective. When optimising, reliable and timely data is absolutely necessary. It is important to ensure that we have got the latest snapshot of what assets are available, without time lags, especially if assets are being actively traded.
Additionally, we need to make sure that we are not stepping on the front office’s toes. Some of the things that we have developed have allowed us to ensure that if an asset manager knows that they are going to be trading a set of assets over a period of, say a week, that we can automatically remove them from optimisation selection, and take them off the table for a period of time. Clients are starting to think about collateral much more from a front office perspective, and it has certainly been spoken about for a number of years—collateral management moving out of the back office towards the front office. Decisions are starting to be made around “what’s the impact to my margin requirements by doing this particular trade or clearing via this central counterparty?” so, it is about looking at the bigger picture and the impacts to performance that collateral can have.
Have regulatory changes challenged the way firms think about collateral management?
The increasing regulatory requirements have added complexity and that is certainly bringing collateral management to the forefront of people’s minds. The decisions about where you might clear a trade can make an impact on the netting benefits at a clearinghouse. Consequently, we are starting to see more scenario analysis tools available, certainly, that’s where we want to get with this tool.
Right now it sits within our organisation, but in future phases, we are not ruling out being able to put it on clients’ desktops so that they can actually run scenario analysis across their book. We are not there yet but that is an aspiration for sure. Some of the clearing members of OTC clearing house offer tools, it is essentially risk scenario analysis but looking at margin implications. The un-cleared margin rules last year were a big impact and transformational for a lot of clients that hadn’t had to previously exchange margin but suddenly found themselves needing to.
We have spoken to a lot of firms that never really had to consider the fact that they might need to hold cash or collateral back for margin. It’s about understanding the impact of a certain trade on your collateral requirements, and then what is the cost of that, particularly if you have got to access cash financing markets and how does that cost impact overall performance.
Have you had positive feedback on the solutions from clients?
We are seeing really encouraging results in the reduction of market value that we are delivering. We have revisited some clients margin portfolios once they have been live for a few months and we have seen some positive numbers. Client feedback has been very positive and those who have started on the path of rolling out want to get the rest of their funds optimised as soon as possible, so we are now working through a pipeline of clients to get them on to the tool.
How do you see the collateral landscape changing over the next five years?
The theme of the increased complexity is not going to go away as there are further regulatory challenges ahead. The new initial margin rules are going to kick in over the next few years and will impact our clients, so I think that there is a lot going on in the industry right now where we are looking to increase messaging automation, and that is the other aspect of efficiency.
I look into efficiency in two key areas. One is an efficient use of the assets that you’ve got, which we call optimisation and the other is infrastructure and messaging, which is the efficiency and the speed at which we can move assets around to cover margin calls. There is a lot of exciting stuff going on around that. I hope that the industry will continue to adopt some of these because standardisation and streamlining is clearly a good thing.
The theme around liquidity is going to continue to grow. We have got clients that sit on both sides of the fence in terms of some having a potential cash shortfall, whereas there are others that are long cash. Interesting models like peer-to-peer lending are becoming more important and we are starting to see more opportunities emerging.
That will continue and, of course, with these trades requiring margining, that trend of increasing complexity will increase. I have spoken about centralisation, and I think that will clearly continue. Modelling from a front-office perspective will become more prevalent as will the tools used to help model the different scenarios around what a specific trade does and what that means from a marginal cost perspective, what that does to your profit and loss, and almost having a front office role in terms of the collateral management function alongside investment management. We certainly see liquidity management and collateral management being two areas that sit pretty neatly together, and that is how we think about some of the products that we look to offer.
NO FEE, NO RISK
100% ON RETURNS If you invest in only one asset servicing news source this year, make sure it is your free subscription to Asset Servicing Times
100% ON RETURNS If you invest in only one asset servicing news source this year, make sure it is your free subscription to Asset Servicing Times
