Pledge GMSLA Anatomy: Difference between revisions
Amwelladmin (talk | contribs) No edit summary |
Amwelladmin (talk | contribs) No edit summary |
||
| (15 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
{{a| | {{a|pgmsla|{{subtable|'''{{pgmsla}}'''<br>{{2018 GMSLA TOC}}}}}}===What is the [[Pledge GMSLA]]?=== | ||
The elegantly titled [[Global Master Securities Lending Agreement (Security Interest over Collateral 2018 version)]] — known to the friends it has rapidly made in the industry as the “'''[[Pledge GMSLA]]'''” is a version of the [[GMSLA]] published in November 2018 and designed exclusively for [[Agent lender|agent lending]] arrangements. Instead of posting [[collateral]] by [[title transfer]], the {{pgmslaprov|Borrower}} [[pledge]]s it. The {{pgmslaprov|Lender}} has a security interest over the collateral, but no right to [[reuse]] or otherwise deal with it. | |||
{{ | |||
{{ | |||
{{pgmslaprov| | |||
=== | ===What’s it for?=== | ||
[[Agent lending]] arrangements where {{pgmslaprov|collateral}} is held in a [[Tri-party collateral arrangement|tri-party collateral system]]. The {{pgmsla}} is designed to reduce the {{pgmslaprov|borrower}}’s [[LRD]] charges, in a nutshell. | |||
When you borrow securities under a [[GMSLA|stock lending agreement]], you tend to over-collateralise—perhaps you give 105 in value of collateral for 100 of securities you have borrowed. This leaves you in the unusual position of being, net, a ''creditor'' to your lender: your lender has an obligation to title transfer the collateral back to you. If it is bust it cannot, and even after you apply close out netting, you’re in the hole to the tune of 5. | |||
With me? | With me? | ||
Now, if your lender is of dubious repute, from a credit perspective, you might have to hold capital against that credit exposure. Okay, | Now, if your lender is of dubious repute, from a credit perspective, you might have to hold capital against that credit exposure. Okay, it’s only 5, but when you’re a bank you do this in big size and it can add up. If, somehow, you can isolate the lender’s credit exposure it is worth doing. | ||
In most cases, you | In most cases, you can’t: most lenders will want to use your collateral in their own operations (to defray the lending costs of lending the securities to you, right?). If they do this then the collateral is gone, and you have no choice but to be a creditor. | ||
Agent lenders are one class of lender who | Agent lenders are one class of lender who isn’t so bothered about reusing the collateral, because ''it'' didn’t lend to you in the first place, but lent its client’s securities to you, and these clients aren’t so bothered about reuse. | ||
=== | ===(Agent) {{pgmslaprov|Lender}}s who might like to use the Pledge GMSLA=== | ||
A GMSLA would be useful and interesting in the following circumstances: | A pledge GMSLA would be useful and interesting in the following circumstances: | ||
*Where the {{ | *'''FI Borrower''': Where the {{pgmslaprov|Borrower}} is a [[bank]] or [[financial institution]] that would incur a capital/balance sheet charge under [[Basel III|Basel]] rules for the return of excess collateral it has provided by [[title transfer]] | ||
*Where the {{ | *'''''Non''-FI Lender''' Where the {{pgmslaprov|Lender}} is ''not'' a [[financial institution]], but rather is owner of long assets which it is seeking to enhance yield on, where its only concern is [[credit mitigation]] and not funding, and it does not therefore need to [[reuse]] the collateral, being happy for it to be “dead-ended” in a [[Tri-party collateral arrangement|tri-party collateral management system]], as long as it is properly— ahh, ''[[perfect]]ly'' — [[pledge]]d, so that should the {{pgmslaprov|Borrower}} default, the lender has recourse to the collateral. | ||
In other words this is likely to be restricted to [[agent lender]]s and quasi-agent lenders ([[Luxembourg fiduciary|fiduciaries]], [[Espievie|espievies]], [[Repackaging programme|repackaging]] vehicles). | In other words this is likely to be restricted to [[agent lender]]s and quasi-agent lenders ([[Luxembourg fiduciary|fiduciaries]], [[Espievie|espievies]], [[Repackaging programme|repackaging]] vehicles). | ||
==={{gmslaprov|Lender}}s who are ''not'' likely to use the Pledge GMSLA=== | |||
Any normal market participants when trading with each other, where the name of the game is ''funding optimisation'' and ''collateral efficiency''. Any securities lender who needs to use, reuse, rehypothecate posted collateral in their operations So brokers, dealers, banks, credit institutions — anyone who cares about balance sheet and capital efficiency — will ''not'' want to take collateral by pledge. | |||
===No-one<ref>''Almost'' no-one.</ref> needs ''both''=== | |||
It is a well-known rule of thumb that any institution with more than one type of the same master agreement will have all kinds of of operational and booking issues, because, systems not being artificially intelligent, there is no way of knowing which of the master agreements to book a given trade to. The good news is that there should be no “use case” for the ''same'' Lender to have both a title transfer ''and'' a pledge GMSLA. Lenders either care about optimising their collateral and funding — most normal market participants do — and they’ll be under a normal GMSLA, or they don’t — you are like some ultra high net worth asset management client, or a [[sovereign wealth fund]] or something, and you have funding literally coming out of your ears, which is why you are in the agent lending programme in the first place. You will be fine with a pledge GMSLA.<ref>Well: unless you are a [[UCITS]] fund, because the pledge doesn’t sufficiently isolate your credit risk to the lender the way title transfer does.</ref> | |||
There are, apparently, ''some'' banks who lend and take collateral through triparty, as if they were principals of an agent lender. I am not sure ''why'' they do that, but they do. | |||
===Major changes=== | ===Major changes=== | ||
*No concept of {{pgmslaprov|Equivalent}} {{pgmslaprov|Collateral}}, seeing as collateral is pledged and dead-ended, so you ''do'' get back what you pledged (and in fact never technically give it away) — there is none of this fuss around [[true sale]] that you have with [[title transfer]] (in that | *No concept of {{pgmslaprov|Equivalent}} {{pgmslaprov|Collateral}}, seeing as collateral is pledged and dead-ended, so you ''do'' get back what you pledged (and in fact never technically give it away) — there is none of this fuss around [[true sale]] that you have with [[title transfer]] (in that there’s no [[recharacterisation]] to a [[secured loan]]: we’re saying it ''is'' a [[secured loan]]). | ||
*Close out works quite differently. | *Close out works quite differently. | ||
{{sa}} | |||
*The normal, 2010, [[title transfer]] [[GMSLA Anatomy]] - a lot more information about stock lending generally there. | |||
{{ref}} | |||
Latest revision as of 13:30, 14 August 2024
What is the Pledge GMSLA?
The elegantly titled Global Master Securities Lending Agreement (Security Interest over Collateral 2018 version) — known to the friends it has rapidly made in the industry as the “Pledge GMSLA” is a version of the GMSLA published in November 2018 and designed exclusively for agent lending arrangements. Instead of posting collateral by title transfer, the Borrower pledges it. The Lender has a security interest over the collateral, but no right to reuse or otherwise deal with it.
What’s it for?
Agent lending arrangements where collateral is held in a tri-party collateral system. The 2018 Pledge GMSLA is designed to reduce the borrower’s LRD charges, in a nutshell.
When you borrow securities under a stock lending agreement, you tend to over-collateralise—perhaps you give 105 in value of collateral for 100 of securities you have borrowed. This leaves you in the unusual position of being, net, a creditor to your lender: your lender has an obligation to title transfer the collateral back to you. If it is bust it cannot, and even after you apply close out netting, you’re in the hole to the tune of 5.
With me?
Now, if your lender is of dubious repute, from a credit perspective, you might have to hold capital against that credit exposure. Okay, it’s only 5, but when you’re a bank you do this in big size and it can add up. If, somehow, you can isolate the lender’s credit exposure it is worth doing.
In most cases, you can’t: most lenders will want to use your collateral in their own operations (to defray the lending costs of lending the securities to you, right?). If they do this then the collateral is gone, and you have no choice but to be a creditor.
Agent lenders are one class of lender who isn’t so bothered about reusing the collateral, because it didn’t lend to you in the first place, but lent its client’s securities to you, and these clients aren’t so bothered about reuse.
(Agent) Lenders who might like to use the Pledge GMSLA
A pledge GMSLA would be useful and interesting in the following circumstances:
- FI Borrower: Where the Borrower is a bank or financial institution that would incur a capital/balance sheet charge under Basel rules for the return of excess collateral it has provided by title transfer
- Non-FI Lender Where the Lender is not a financial institution, but rather is owner of long assets which it is seeking to enhance yield on, where its only concern is credit mitigation and not funding, and it does not therefore need to reuse the collateral, being happy for it to be “dead-ended” in a tri-party collateral management system, as long as it is properly— ahh, perfectly — pledged, so that should the Borrower default, the lender has recourse to the collateral.
In other words this is likely to be restricted to agent lenders and quasi-agent lenders (fiduciaries, espievies, repackaging vehicles).
Lenders who are not likely to use the Pledge GMSLA
Any normal market participants when trading with each other, where the name of the game is funding optimisation and collateral efficiency. Any securities lender who needs to use, reuse, rehypothecate posted collateral in their operations So brokers, dealers, banks, credit institutions — anyone who cares about balance sheet and capital efficiency — will not want to take collateral by pledge.
No-one[1] needs both
It is a well-known rule of thumb that any institution with more than one type of the same master agreement will have all kinds of of operational and booking issues, because, systems not being artificially intelligent, there is no way of knowing which of the master agreements to book a given trade to. The good news is that there should be no “use case” for the same Lender to have both a title transfer and a pledge GMSLA. Lenders either care about optimising their collateral and funding — most normal market participants do — and they’ll be under a normal GMSLA, or they don’t — you are like some ultra high net worth asset management client, or a sovereign wealth fund or something, and you have funding literally coming out of your ears, which is why you are in the agent lending programme in the first place. You will be fine with a pledge GMSLA.[2]
There are, apparently, some banks who lend and take collateral through triparty, as if they were principals of an agent lender. I am not sure why they do that, but they do.
Major changes
- No concept of Equivalent Collateral, seeing as collateral is pledged and dead-ended, so you do get back what you pledged (and in fact never technically give it away) — there is none of this fuss around true sale that you have with title transfer (in that there’s no recharacterisation to a secured loan: we’re saying it is a secured loan).
- Close out works quite differently.
See also
- The normal, 2010, title transfer GMSLA Anatomy - a lot more information about stock lending generally there.